Surgical treatment of congenital IHPSSs in dogs represents one of the greatest challenges faced by veterinary surgeons. Complete attenuation of portosystemic shunts is recommended to achieve resolution of clinical signs and improve long-term survival.1,2 However, an IHPSS typically has large anomalous vessels that are difficult to access because of their location within the hepatic parenchyma.3–5 When acute complete attenuation of IHPSS is achievable, intolerable portal hypertension occurs in most affected dogs.6 A variety of surgical techniques have been proposed to achieve progressive attenuation of portosystemic shunts,7 including partial occlusion with silk,8 ameroid constrictors,6 cellophane banding,9 and transvenous coil embolization.a Although each of these techniques has resulted in successful attenuation of IHPSS in selected dogs, no technique achieves safe and reliable occlusion of IHPSS after a single surgical intervention. Preliminary evaluation of a silicone HO in a rodent model of vascular occlusion revealed a gradual decrease in blood flow that would be desirable for occlusion of an IHPSS.10 Subsequent laboratory evaluation suggested that the device was suitable for long-term maintenance of occlusion.11 The purpose of the present study was to evaluate efficacy and complications associated with use of a percutaneously controlled silicone HO for gradual occlusion of IHPSSs in dogs. Our hypothesis was that the HO would provide an effective method for complete occlusion of an IHPSS with minimal risk of perioperative portal hypertension.
Materials and Methods
Dogs—A prospective clinical study was performed that included all dogs at the University of Florida Veterinary Medical Center treated for a single IHPSS from January 2004 through February 2005. All procedures were approved by the Institutional Animal Care and Use Committee, and clients signed an informed consent form that was reviewed by the Clinical Research Review Committee at the University of Florida. Standard medical management with low-protein diet, lactulose (5 to 15 mL, PO, q 8 to 12 h), and an orally administered antimicrobial effective against anaerobic bacteria (neomycin, 20 mg/kg [9.1 mg/lb], PO, q 8 to 12 h) was instituted for at least 1 week prior to surgery. Medical management was continued after surgery, until confirmation of shunt occlusion. Prior to surgery, abdominal ultrasonography, CBC, serum biochemical analyses, determination of serum bile acids concentration (after withholding food for 12 hours and 2 hours after eating), and transcolonic scintigraphy (technetium Tc 99M pertechnate) were performed on all dogs.
Implants—Heavy-duty silicone hydraulic vascular occluders were modified by the manufacturerb at the request of the authors, decreasing the width of the cuff from 10 mm to 6 mm and tapering the suture eyelets to minimize the tissue dissection required for placement of the occluder around the portal vasculature (Figure 1). Unmodified subcutaneous injection portsc were obtained from a separate manufacturer. All implants were sterilized by use of ethylene oxide. Prior to implantation, air was eliminated from the HO lumen by filling it with sterile saline (0.9% NaCl) solution in a retrograde manner with a 19-gauge IV catheter.d The injection port was similarly filled with saline solution by use of a Huber point needlee and 3-mL syringe, then connected to the actuating tubing of the HO. The HO was exercised and leak tested by inflating to complete occlusion 5 times.10 After preparation, the volume of saline solution required for complete occlusion of the HO was recorded to ± 0.1 mL and used to plan an injection schedule for staged occlusion following surgery.
Photograph of a modified HO (side view) used for treatment of dogs with IHPSSs. The width of the cuff was decreased from 10 to 6 mm, and the suture eyelets (white arrow) were tapered to allow for easier passage around a thin vessel.
Citation: Journal of the American Veterinary Medical Association 229, 11; 10.2460/javma.229.11.1749
Photograph of a modified HO (side view) used for treatment of dogs with IHPSSs. The width of the cuff was decreased from 10 to 6 mm, and the suture eyelets (white arrow) were tapered to allow for easier passage around a thin vessel.
Citation: Journal of the American Veterinary Medical Association 229, 11; 10.2460/javma.229.11.1749
Photograph of a modified HO (side view) used for treatment of dogs with IHPSSs. The width of the cuff was decreased from 10 to 6 mm, and the suture eyelets (white arrow) were tapered to allow for easier passage around a thin vessel.
Citation: Journal of the American Veterinary Medical Association 229, 11; 10.2460/javma.229.11.1749
Surgical procedures—The liver was accessed through a standard ventral midline approach. Anatomic localization of the shunt was confirmed by prehepatic dissection and measurement of portal venous pressures during complete occlusion of the portal vein branch leading to the IHPSS. Portal venous pressure measurements were repeated after the uninflated HO was placed around the portal vein branch and were later compared with baseline portal pressures by use of 1-way ANOVA. The luminal diameter of the uncompressed vessel was estimated with a metric ruler, and an HO of similar lumen diameter was selected to avoid compression of the vessel by the uninflated cuff at the time of application. A strand of size 0 polypropylene suture was passed around the portal vein branch leading to the IHPSS and threaded through the suture eyelets of the HO. The HO was pulled around the portal vein branch by use of the suture and digital manipulation (Figure 2). Passage of the HO around the portal vein branch was facilitated by coating the occluder with sterile methylcellulose lubricant. Suture eyelets were then apposed with a single ligature of size 0 polypropylene suture, enclosing the vessel within the HO. Actuating tubing from the HO was passed to the exterior through a separate stab incision in the paramedian portion of the abdominal wall and connected to a subcutaneous injection port that was placed 5 cm lateral to the abdominal midline and 3 to 5 cm caudal to the ribs. In the first 3 dogs, the actuating tubing was affixed to the subcutaneous port with 2 circumferential sutures of 3-0 polypropylene. After detachment of the actuating tubing from the port occurred in 1 dog, the actuating tubing was affixed to the port by use of a plastic boot connector supplied by the manufacturer (Figure 3).f Routine closure of the abdominal and skin incisions was performed, and total operative time from skin incision to closure was recorded on the anesthetic record.
Intraoperative photographs of surgical placement of an HO in a dog. A—Initial positioning of the HO (black arrowhead) and actuating tubing (white arrow) around the right branch of the portal vein (black arrow) in a dog with a right divisional intrahepatic shunt. B—Completed placement of the HO (white arrowhead) around the right caudal branch (white arrow) of the portal vein (PV) in a dog. A length of polypropylene suture was placed through the suture eyelets and tied, then marked with hemoclips for later identification. GB = Gallbladder.
Citation: Journal of the American Veterinary Medical Association 229, 11; 10.2460/javma.229.11.1749
Intraoperative photographs of surgical placement of an HO in a dog. A—Initial positioning of the HO (black arrowhead) and actuating tubing (white arrow) around the right branch of the portal vein (black arrow) in a dog with a right divisional intrahepatic shunt. B—Completed placement of the HO (white arrowhead) around the right caudal branch (white arrow) of the portal vein (PV) in a dog. A length of polypropylene suture was placed through the suture eyelets and tied, then marked with hemoclips for later identification. GB = Gallbladder.
Citation: Journal of the American Veterinary Medical Association 229, 11; 10.2460/javma.229.11.1749
Intraoperative photographs of surgical placement of an HO in a dog. A—Initial positioning of the HO (black arrowhead) and actuating tubing (white arrow) around the right branch of the portal vein (black arrow) in a dog with a right divisional intrahepatic shunt. B—Completed placement of the HO (white arrowhead) around the right caudal branch (white arrow) of the portal vein (PV) in a dog. A length of polypropylene suture was placed through the suture eyelets and tied, then marked with hemoclips for later identification. GB = Gallbladder.
Citation: Journal of the American Veterinary Medical Association 229, 11; 10.2460/javma.229.11.1749
Photograph of the actuating tubing of a silicone HO attached to the subcutaneous injection port with a boot connector (white arrow) to minimize the risk of detachment.
Citation: Journal of the American Veterinary Medical Association 229, 11; 10.2460/javma.229.11.1749
Photograph of the actuating tubing of a silicone HO attached to the subcutaneous injection port with a boot connector (white arrow) to minimize the risk of detachment.
Citation: Journal of the American Veterinary Medical Association 229, 11; 10.2460/javma.229.11.1749
Photograph of the actuating tubing of a silicone HO attached to the subcutaneous injection port with a boot connector (white arrow) to minimize the risk of detachment.
Citation: Journal of the American Veterinary Medical Association 229, 11; 10.2460/javma.229.11.1749
Gradual occlusion of the HO—Dogs were reevaluated at 2, 4, 6, and 8 weeks after surgery, and injections of sterile saline solution were made into the subcutaneous injection ports to achieve gradual inflation of the HO. Prior to each injection, dogs were positioned in lateral recumbency, and the area over the injection port was shaved and aseptically prepared. A 22-gauge, right-angle Huber needle extension sete was used to penetrate the skin and underlying injection port, and sterile saline solution was injected to achieve 25% of the filling volume at 2 weeks, 50% at 4 weeks, and 75% at 6 weeks. On the basis of results of a previous study10 that used the HO, an additional volume of sterile saline solution was injected to achieve a total of 120% of the original filling volume at the 8-week recheck visit. Dogs were kept in the hospital for 2 to 4 hours to allow observation for clinical signs of acute portal hypertension (shock, hematochezia, vomiting, abdominal distension), then discharged to the owners.
Outcome assessment—Short-term outcome was assessed 2 weeks after completion of the injection schedule; all previous tests were repeated. Continuous data were compared between preoperative and 2-week postocclusion time points by use of repeated-measures ANOVA. Complications were recorded and classified as intraoperative, postoperative, or implant related. Dogs were reevaluated at the University of Florida or by their local veterinarian at 1 year following surgery by evaluation of postprandial bile acids concentration and serum biochemical panels.
Results
Dogs—10 dogs were entered into the study. Breeds represented were Labrador Retriever (n = 3 dogs), Golden Retriever (3), Coonhound (1), Welsh Pembroke Corgi (1), Doberman Pinscher (1), and mixed breed (1). Median age at the time of surgery was 6 months (range, 4 to 24 months).
Preoperative diagnostic findings—Preoperative diagnosis of a single IHPSS was confirmed in all 10 dogs on the basis of results of pre- and postprandial bile acids determinations and identification of a single intrahepatic shunt via abdominal ultrasonographic examination. Detection of portosystemic shunting by use of transcolonic scintigraphy was obtained in 8 of 10 dogs, whereas results were nondiagnostic in 2 dogs because of poor absorption of the radioisotope.
Surgical procedure—In all dogs, adequate access to the liver and surrounding vasculature was achieved through a midline abdominal incision; however, a caudal median sternotomy was performed in 1 dog to confirm identification of the vena cava and azygous vein when situs inversus viscerum was discovered during initial abdominal exploration. In all dogs, the portal vein branch leading to the IHPSS was identified by dissection of the porta hepatis and identification of an atypically large vessel. Prehepatic isolation and temporary occlusion of the portal branch leading to the IHPSS led to increases in portal pressure > 22 cm H2O in all 10 dogs. On the basis of the scheme proposed by Lamb and White,12 anatomic shunt location was classified as right divisional in 6 dogs, left divisional in 3 dogs, and central divisional in 1 dog. Mean ± SD portal pressure prior to occlusion (8.0 ± 3.1 cm H2O) was not significantly different from portal pressure measured after placement of the HO (9.2 ± 3.7 cm H2O). Mean duration of surgery (skin incision to skin closure) was 192 ± 32 minutes.
Intraoperative complications occurred in 2 dogs and included hemorrhage during dissection (n = 1) and severe hypotension and oliguric renal failure (1). In the first dog, a 3-mm tear in the vena cava was located and was closed with 6-0 polypropylene suture in a simple continuous pattern. The second dog was treated for acute renal failure with furosemide (2 to 6 mg/kg [0.91 to 2.7 mg/lb], IV, q 2 to 6 h) and the gastric protectant famotidine (0.5 mg/kg [0.23 mg/lb], IV, q 12 h). Bicarbonate dose was titrated to control hyperkalemia and acidosis (3.5 mEq/kg/h [1.6 mEq/lb/h]), and urine output and insensible water losses were used to calculate IV fluid rate. Urine production returned at 48 hours after surgery, and azotemia resolved completely within 5 days. All dogs survived surgery and were discharged from the hospital.
Postoperative complications—Implant revision was required in 3 of the first 6 dogs during the inflation period. In 2 dogs, mechanical failure of the HOs was suspected because of loss of internal pressure at the time of the final saline solution injection. Sterile, water-soluble contrast material (meglumine diatrizoate) was injected into the subcutaneous port, and abdominal radiography was used to detect intraabdominal leakage from the inflatable cuff. Both HOs were replaced during revision surgeries, inflation of the occluders was immediately returned to 50% at the time of surgery, and dogs were returned to the previous injection schedule without further complication. Examination of the failed implants revealed rupture at a silicone seam. The manufacturer was contacted, and seams were reinforced in subsequent implants. The complication did not occur in the final 4 dogs. In 1 dog, detachment of the port from the actuating tubing occurred and was revised with a plastic bootf as described. Evaluation of serum biochemical values and postprandial bile acids concentrations indicated continued portosystemic shunting in all 3 dogs that had implant deflation, indicating that the deflated device did not cause occlusion by inflammation or fibrosis over this 10-week period. An incisional infection was diagnosed in 1 dog on the basis of fever (rectal temperature, 38.5°C [103.1°F]) and serosanguinous discharge from the cranial aspect of the incision. Signs resolved following placement of an Elizabethan collar and administration of amoxicillin-clavulanic acid (15 mg/kg [6.8 mg/lb)], PO, q 12 h for 21 days). Two dogs developed ascites after surgery. Ascites resolved without specific treatment in 1 dog within 7 days of surgery, and complete inflation of the HO was performed without complication. In the other dog, ascites developed during a period of acute oliguric renal failure and persisted until the 2-week recheck evaluation. Inflation of the occluder was delayed for another 2 weeks to avoid exacerbation of ascites. By the third week after surgery, ascites resolved spontaneously, and inflation of the occluder was performed without complication. On the basis of postprandial bile acids concentrations, both dogs had normal liver function at final follow-up, although results of scintigraphy suggested continued portosystemic shunting in 1 dog.
Short-term follow-up—All 10 dogs survived through the short-term (10-week) follow-up period. Results of serum biochemical analyses indicated significant improvements in serum albumin (P < 0.001), total cholesterol (P < 0.001), and BUN (P = 0.016) concentrations (Table 1). Postprandial bile acids concentrations within reference range were detected in 6 of 10 dogs at short-term follow-up, and 1 additional dog had postprandial bile acids concentration slightly greater than the reference range. In the 3 dogs with most severe abnormalities in postoperative serum bile acids concentrations, values improved significantly (P = 0.04) from preoperative (247 ± 45 mmol/L) to postoperative values (119 ± 58 mmol/L). Results of scintigraphy suggested that, despite the fact that postprandial bile acids concentrations returned to reference range in 6 of 10 dogs, portosystemic shunting continued in 5 of 10 dogs and was not well correlated with serum bile acids concentrations. At short-term follow-up, 2 dogs had negative results of scintigraphy concurrent with abnormal postprandial bile acids concentrations, whereas 3 dogs had positive results of technetium scans concurrent with postprandial bile acids concentrations that were within reference range. Nonselective computed tomographic angiography13 was performed in 2 of these dogs, and results suggested incomplete occlusion of the HO. No evidence of multiple extrahepatic shunts was detected in either dog. No clinical signs of hepatic encephalopathy were detected in 9 of 10 dogs. One dog had generalized seizures prior to surgery that continued after occlusion of the HO, despite having postprandial bile acids and serum ammonia concentrations within reference ranges and no scintigraphic evidence of portosystemic shunting. Magnetic resonance imaging of the brain revealed cortical atrophy, but no specific lesions were discovered. Results of CSF examination were unremarkable. The dog is being treated with anticonvulsants (phenobarbital, 2 mg/kg [0.91 mg/lb], PO, q 12 h; potassium bromide, 20 mg/kg, PO, q 12 h) and is clinically normal except for the seizure disorder.
Selected serum biochemical and liver function test results (mean ±SD) obtained prior to surgery, 2 weeks after complete inflation of an HO, and 12 months after surgery In dogs with IHPPSs.
| Time of sampling | No.of dogs | BUN (mg/dL) | Albumin (g/dL) | TP (g/dL) | Cholest (mg/dL) | Glucose (mg/dL) | Bilirubin (mg/dL) | PPSBA (mMol/L) |
|---|---|---|---|---|---|---|---|---|
| Preoperative | 10 | 5.3 ± 2.1 | 2.1 ± 0.29 | 4.4 ± 0.68 | 187 ± 71 | 106 ± 24 | 0.1 ± 0.1 | 228 ± 102 |
| After occlusion (2 wk) | 10 | 8.3 ± 4.2 | 2.9 ± 0.26 | 5.7 ± 0.49 | 298 ± 62 | 99 ± 8.3 | 0.1 ± 0.1 | 43 ± 60 |
| Follow-up (1 y) | 8 | 13.1 ± 7.9 | 3.3 ± 0.4 | 6.1 ± 0.7 | 285 ± 62 | 106 ± 10 | 0.2 ± 0.1 | 38.5 ± 54.1 |
| Reference range | 7–25 | 2.8–3.8 | 5.8–7.8 | 133–367 | 70–122 | 0–0.4 | 0–25 |
PPSBA = Postprandial serum bile acids concentration. TP = Total protein. NA = Not applicable.
Last follow-up—Follow up was performed at 12 months after surgery by direct examination at the Veterinary Medical Center (n = 6) or by phone conversation with the primary veterinarian (4) for all dogs. Two dogs were lost to long-term follow-up: 1 client moved 6 months after surgery without leaving forwarding information, and a second dog died of unrelated causes (foreign body ingestion and septic peritonitis) 4 months after HO placement. Both dogs had postprandial bile acids concentrations within reference range at the time of last follow-up. In the remaining 8 dogs, postprandial bile acids concentrations were within reference range in 5 dogs at 12 months after surgery. One dog that had been receiving phenobarbital had postprandial bile acids concentration within reference range at the time of short-term follow-up, but had slightly high postprandial bile acids concentration at the 1-year follow-up. Another dog had high postprandial bile acids concentration at the short-term followup, but an additional injection of saline solution was performed 8 months after surgery, and bile acids concentration was within reference range at the 1-year recheck visit. Overall, 8 of 10 dogs achieved postprandial bile acids concentrations within reference range at some time after surgery. In the 8 dogs available for long-term follow-up, all were clinically normal at a median of 22 months (range, 13 to 28 months) after surgery, and no dog is known to have died of the primary disease.
One dog developed a fistulous tract associated with the subcutaneous injection port 9 months after surgery. The fistulous tract resolved with antimicrobial administration (cephalexin, 20 mg/kg, PO, q 12 h) but recurred after administration ceased. The subcutaneous port and actuating tubing were removed 12 months after surgery, but have been retained in all other dogs to provide access for reinflation of the occluder, if required.
Discussion
Use of the HO appears to have potential as a safe and effective method for surgical treatment of dogs with IHPSS. No perioperative deaths occurred in the 10 dogs in this study, which compared favorably with reported perioperative mortality rates for alternative conventional extravascular surgical techniques such as cellophane banding (27%),9 jugular venograft and ameroid occluder placement (10%),6 and partial occlusion with silk (23%, 11%, and 18%).8,14–16 Similar to cellophane banding, ameroid constrictors, and silk suture, the HO is relatively inexpensive and may be applied in any veterinary referral center without the use of specialized equipment.
Previous studies have used a variety of methods for outcome assessment following treatment of dogs with IHPSS, including evaluation of clinical signs14,15 and serum activities of liver enzymes4,6 and technetium scintigraphy.6 Results of the present study were consistent with those of previous studies,6,9,14,15 with resolution of clinical signs in most dogs that survived the perioperative period. Although resolution of clinical signs is the ultimate goal for any treatment method, results of the present study indicated that clinical improvement and standard serum biochemical variables (eg, albumin, cholesterol, and BUN) are relatively insensitive tests for postoperative assessment of liver function, compared with postprandial bile acids concentration and scintigraphy. All dogs in the present study had resolution of clinical signs related to the portosystemic shunt, and all dogs had improvements in serum biochemical variables after surgery; however, only 6 of 10 dogs had postprandial bile acids concentrations within reference range, and 5 of 10 had continued portosystemic shunting detected by use of scintigraphy 2 weeks after complete occlusion.
The lack of correlation between postprandial bile acids concentrations and scintigraphy is difficult to explain. High serum bile acids concentration is a highly sensitive indicator of liver dysfunction, with a sensitivity of 100% when a cutoff of 25 μmol/L is used.17 Most dogs with portosystemic shunts have postprandial bile acids concentrations > 100 μmol/L.18 Although high bile acids concentrations are not correlated with severity of liver dysfunction among affected dogs, changes over time may be used to assess clinical improvements in individual dogs.19 Conversely, transcolonic scintigraphy has been considered the gold standard for preoperative noninvasive confirmation of portosystemic shunting, with reported sensitivity and specificity of 98% and 100%.20 Although the use of scintigraphy is well established in the preoperative diagnosis of portosystemic shunting, the sensitivity and specificity of postoperative transcolonic scintigraphy have not been evaluated in dogs. In addition, there is much debate regarding the rate of occlusion induced by each of the implants,21,22 making it unclear when it is most appropriate to perform scintigraphy to obtain results that are predictive of long-term outcome for each of the vascular occlusion techniques. At 2 weeks after occlusion, 3 dogs in the present study had positive results of technetium scans concurrent with postprandial bile acids concentrations within reference range. It is possible that a small degree of portosystemic shunting may have persisted in some dogs in the present study at a level that was detectable via scintigraphy but did not lead to abnormalities in liver function tests. Conversely, 2 dogs had negative results of scintigraphy concurrent with abnormal postprandial bile acids concentrations. Interestingly, both of these dogs had bile acids concentrations within reference range 1 year after surgery. Ideally, it would have been informative to have repeated the scintigraphy at the 1-year follow-up visit to obtain information on the relationships among clinical signs, liver enzyme activities, and persistence of portosystemic shunting. Correlations between serum biochemical variables, serum bile acids concentrations, or results of scintigraphy and long-term outcome remain to be determined in affected dogs.
In a previous study6 with short-term follow-up, a high prevalence of persistent shunting caused by formation of multiple extrahepatic shunts after surgical treatment of IHPSS with a jugular venograft and ameroid occluder was reported, suggesting either premature occlusion induced by the ameroid constrictor or thrombosis of the vascular graft. In that study, transcolonic scintigraphy revealed continued portosystemic shunting in 5 of 8 dogs, with multiple extrahepatic shunts confirmed in 4 of 5 dogs that were examined via portography or exploratory surgery. The prevalence of multiple extrahepatic shunt formation after application of the HO was difficult to discern on the basis of the available data for the present study. After application of the HO, the rate of persistent portosystemic shunting detected via scintigraphy 2 weeks after complete occlusion (50%) was similar to that reported by Kyles et al (62.5%),6 although serum biochemical variables and serum bile acids concentrations were not reported for that study. Unfortunately, it was difficult to recommend invasive postoperative testing for multiple extrahepatic shunts in the present study because clinical signs and, in many instances, liver enzyme activities had returned to reference ranges.
The theoretical advantage of the HO over materials used previously in the surgical treatment of dogs with IHPSS is the ability to perform postoperative adjustments in the rate of vascular occlusion. Vascular occlusion occurs after partial occlusion with silk suture, ameroid constrictors, or cellophane and is controlled primarily by endogenous factors such as inflammation and thrombosis. On the basis of the high rate of formation of multiple extrahepatic shunts in previous studies of the use of ameroid constrictors for occlusion of IHPSS,6 an experimental study21 of the rate of venous occlusion induced by various devices used in the treatment of dogs with portosystemic shunts, and a previous report8 of successful staged occlusion of portosystemic shunts at 4 to 6 weeks after the initial surgery, we used a conservative timetable for inflation of the HO, achieving complete inflation over an 8-week period. The absence of postoperative portal hypertension in the first 10 dogs suggested that it may be possible to inflate the HO at a more rapid rate in future studies. However, except for decreasing the duration of the follow-up period for the dog and client, there does not appear to be a reason to increase the rate of vascular occlusion. Presently, the authors continue to adhere to the injection schedule described here. An unexpected advantage of the HO technique was that it also allowed adjustment of the rate of vascular occlusion for each individual dog. For example, the first injection was delayed in 1 dog that had persistent ascites at the time of the 2-week follow-up examination. In a second dog in which liver enzyme activities did not return to reference ranges after the initial injection series, additional injections led to a return to reference ranges at 1 year after surgery. Arguably, these postoperative adjustments would have been difficult or impossible with other vascular occlusion techniques. Another advantage of the HO is that the silicone implant is malleable and light, with smooth edges that may decrease risk of vascular tearing during placement, although dissection of a space large enough for placement of the device remains technically challenging.
Although outcome for this technique was favorable in comparison to previous reports for animals with IHPSSs, use of the HO occluder also has a number of disadvantages. Device-related complications including rupture of the silicone membrane and detachment of the actuating tubing from the subcutaneous injection port occurred. Fortunately, revision of the method of implant manufacture and use of a new technique for attachment of actuating tubing to the injection port led to resolution of these problems in the last 4 dogs. Since completion of the study, surgery and complete inflation of the modified HO have been performed in 6 dogs without any evidence of implant failure. A long-term complication with the device (a fistulous tract) occurred in only 1 dog and was resolved by removal of the actuating tubing and subcutaneous injection port. Another disadvantage is the requirement for multiple follow-up visits for injections into the subcutaneous injection port. Also, because the HO is applied as an extravascular occlusive device and requires delicate dissection of prehepatic portal vasculature, surgical implantation is not likely to be easily achieved through minimally invasive techniques. Although the occluder could be passed around the portal vein branches in 10 consecutive dogs, the size of the HO makes passage of the device around the target vessel more technically challenging than placement of silk or cellophane bands, which are thinner and more malleable. The prehepatic placement of the HO in this study was strongly influenced by surgeon preference for dissection of the portal vein branches, which readily allows identification of the enlarged branch leading to the shunt and avoids the need to incise the diaphragm and sternum in an attempt to expose the hepatic veins. The high prevalence of right and central divisional shunts may have also contributed to the common choice of prehepatic dissection in this study.
The medical grade silicone used in manufacture of the HO induces minimal inflammation and is similar in composition to permanent silicone implants presently approved for use in humans. On the basis of experience with other methods of vascular occlusion, the authors specifically sought out a device for occlusion of portosystemic shunts that was percutaneously controllable and that would achieve complete luminal occlusion with minimal inflammation, theoretically allowing a more consistent and predictable rate of vascular occlusion than is possible with implants that induce progressive vasculitis and thrombosis without achieving luminal closure. A decrease in inflammation associated with a device for gradual occlusion could have both positive and negative aspects. For example, if silicone induces minimal inflammation, there may be a concern about recanalization of the vessel as the occluder slowly loses pressure over time. Conversely, vascular occlusion induced by inflammation may also be transient, and recanalization can occur as the degree of inflammation decreases over time. Another area of interest is the shape of the vessel during vascular attenuation and the subsequent effects on hemodynamics. The uninflated HO begins as a circular cuff, but compresses the vessel into a flat structure as the cuff folds at a single seam. This may result in different hemodynamics than those induced by a cellophane band or an ameroid constrictor, although a study10 of the HO in a rat caval occlusion model suggests that a gradual decrease in blood flow occurs after staged inflation of the HO. Unfortunately, it is extremely difficult to compare the short- and long-term hemodynamic effects of the various methods of gradual vascular occlusion in a model that adequately simulates the portal vasculature and dynamic hepatic vascular resistance that occur in dogs with naturally occurring portovascular anomalies. In the future, advances in technology may result in a cost-effective implant that may be placed at the time of surgery, allowing surgeons to perform monitoring of blood flow in client-owned dogs with portosystemic shunts during vascular occlusion. For now, techniques are likely to be compared on the basis of ease of application, complications, liver function tests, and long-term control of clinical signs.
ABBREVIATIONS
| HO | Hydraulic occluder |
| IHPSS | Intrahepatic portosystemic shunt |
Weisse C, Solomon JA, Holt D, et al. Percutaneous transvenous coil embolization of canine intrahepatic portosystemic shunts: short term results in 14 dogs (abstr). Vet Surg 2003;32:499–500.
DOCXS Biomedical Products and Accessories, Ukiah, Calif.
Le Grande Access Port, Access Technologies, Skokie, Ill.
22 G × 8-inch BD Intracath central venous catheter, Becton, Dickinson & Co, Franklin Lakes, NJ.
Posi-Grip Huber Point Needle with right angle infusion set, Access Technologies, Skokie, Ill.
Boot connector for 7-F catheter, Access Technologies, Skokie, Ill.
Reference
- 1
Swalec KM, Smeak DD. Partial versus complete attenuation of single portosystemic shunts. Vet Surg 1990;19:406–411.
- 2
Komtebedde J, Koblik PD, Breznock EM, et al. Long-term clinical outcome after partial ligation of single extrahepatic vascular anomalies in 20 dogs. Vet Surg 1995;24:379–383.
- 3
Hunt GB, Bellenger CR, Pearson MRB. Transportal approach for attenuating intrahepatic portosystemic shunts in dogs. Vet Surg 1996;25:300–308.
- 4
Tobias KM, Byarlay JM, Henry RW. A new dissection technique for approach to right-sided intrahepatic portosystemic shunts: anatomic study and use in three dogs. Vet Surg 2004;33:32–39.
- 5
Breznock EM, Berger B, Pendray D, et al. Surgical manipulation of intrahepatic portocaval shunts in dogs. J Am Vet Med Assoc 1983;182:798–805.
- 6↑
Kyles AE, Gregory CR, Jackson J, et al. Evaluation of a portocaval venograft and ameroid ring for the occlusion of intrahepatic portocaval shunts in dogs. Vet Surg 2001;30:161–169.
- 7↑
Sereda CW, Adin CA. Methods of gradual vascular occlusion and their applications in treatment of congenital portosystemic shunts in dogs: a review. Vet Surg 2005;34:83–91.
- 8↑
Van Vechten BJ, Komptebedde J, Koblik PD. Use of transcolonic scintigraphy to monitor blood flow and progressive postoperative attenuation of partially ligated extrahepatic portosystemic shunts in dogs. J Am Vet Med Assoc 1994;204:1770–1774.
- 9↑
Hunt GB, Kummeling A, Tisdall PL, et al. Outcomes of cellophane banding for congenital portosystemic shunts in 106 dogs and 5 cats. Vet Surg 2004;33:25–31.
- 10↑
Sereda CW, Adin CA, Ginn PE, et al. Evaluation of a percutaneously controlled hydraulic occluder in a rat model of gradual venous occlusion. Vet Surg 2005;34:35–42.
- 11↑
Sereda CW, Adin CA, Batich CD, et al. Evaluation of manufacturing variability, diffusion of filling solutions, and long-term maintenance of occlusion in 3 sizes of silicone hydraulic occluders. Am J Vet Res 2006;67:1453–1458.
- 12↑
Lamb CR, White RN. Morphology of congenital intrahepatic portocaval shunts in dogs and cats. Vet Rec 1998;142:55–60.
- 13↑
Frank P, Mahaffey M, Egger C, et al. Helical computed tomographic portography in ten normal dogs and ten dogs with a portosystemic shunt. Vet Radiol Ultrasound 2003;44:392–400.
- 14
Bostwick DR, Twedt DC. Intrahepatic and extrahepatic portal venous anomalies in dogs: 52 cases (1982–1992). J Am Vet Med Assoc 1995;206:1181–1185.
- 15
White RN, Burton CA, McEvoy FJ. Surgical treatment of intrahepatic portosystemic shunts in 45 dogs. Vet Rec 1998;142:358–365.
- 16
Papazoglou LG, Monnet E, Seim HB III. Survival and prognostic indicators for dogs with intrahepatic portosystemic shunts: 32 cases (1990–2000). Vet Surg 2002;31:561–570.
- 17↑
Johnson SE. Chronic hepatic disorders. In:Ettinger ST, Feldman EC, ed.Textbook of veterinary internal medicine. 5th ed.Philadelphia: WB Saunders Co, 2000;1298–1323.
- 18↑
Center SA, ManWarren T, Slater MR, et al. Evaluation of twelve-hour preprandial and two-hour postprandial serum bile acids concentrations for diagnosis of hepatobiliary disease in dogs. J Am Vet Med Assoc 1991;199:217–226.
- 19↑
Bostwick DR, Meyer DJ. Bilirubin and bile acids in the diagnosis of hepatobiliary disease. In:Bonagura JD, ed.Current veterinary therapy (small animal practice) XII. Philadelphia: WB Saunders Co, 1995;736–740.
- 20↑
Koblik PD, Hornoff WJ. Transcolonic sodium pertechnetate Tc 99m scintigraphy for diagnosis of macrovascular portosystemic shunts in dogs, cats and potbellied pigs: 176 cases (1988–1992). J Am Vet Med Assoc 1995;207:729–733.
- 21↑
Youmans KR, Hunt GB. Experimental evaluation of four methods of progressive venous occlusion in dogs. Vet Surg 1999;28:38–47.
- 22
Besancon MF, Kyles AE, Griffey SM, et al. Evaluation of the characteristics of venous occlusion after placement of an ameroid constrictor in dogs. Vet Surg 2004;33:597–605.
