Portal and splanchnic hemodynamics after ... - Wiley Online Library

1 downloads 0 Views 481KB Size Report
partial splenic embolization in cirrhotic patients with hypersplenism. APMIS 2015; 123: 1032–1039. To assess the acute effects of partial splenic embolization ...
© 2015 APMIS. Published by John Wiley & Sons Ltd. DOI 10.1111/apm.12470

APMIS 123: 1032–1039

Portal and splanchnic hemodynamics after partial splenic embolization in cirrhotic patients with hypersplenism AHMED Z. HELALY,1 MOHAMED S. AL-WARRAKY,2 GASSER I. EL-AZAB,3 MOHAMED A. S. KOHLA3 and ELSAYED E. ABDELAAL3 1

Internal Medicine Department, Alexandria Faculty of Medicine, Alexandria; 2Department of Radiology; and 3 Department of Hepatology, National Liver Institute, Menoufiya University, Shebeen El kom, Egypt

Helaly AZ, Al-Warraky MS, El-Azab GI, Kohla MAS, Abdelaal EE. Portal and splanchnic hemodynamics after partial splenic embolization in cirrhotic patients with hypersplenism. APMIS 2015; 123: 1032–1039 To assess the acute effects of partial splenic embolization (PSE) on portal and splanchnic hemodynamics in patients with cirrhosis. Ninety-five patients with hypersplenism were included in the study. Duplex examinations were performed before and 3 and 7 days after PSE. Portal and splanchnic hemodynamics including vessel cross-sectional area (CSA), mean flow velocities (cm/s), blood flows (mL/min), Doppler indices as portal congestion index (CI), liver vascular index, hepatic artery and superior mesenteric artery (SMA) pulsatility and resistive indices (PI and RI), were performed before and after PSE. In our study, 69 of 95 patients were males (72.6%) and 26 females (27.3%). Chronic hepatitis C virus infection was the main cause of cirrhosis (81.1%). PSE failed technically in six patients (6.3%). After PSE, both CSA and CI significantly decreased (p < 0.05 and 0.05).

1035

PORTAL HEMODYNAMICS AFTER PSE

DISCUSSION Bleeding from esophageal varices is a common and life-threatening complication in patients with cirrhosis and portal hypertension. Several approaches to control bleeding from esophageal varices were developed to lower either the resistance (i.e., shunt surgery) or the portal inflow (i.e., beta adrenergic blockers or PSE). The effectiveness of these approaches on portal decompression would be dependent, at least in part, on whether increased resistance or increased inflow is the predominant mechanism for portal hypertension (11). Improving portal hemodynamics is usuaslly associated with a reduction in the sequelae of portal hypertension, particularly, bleeding esophageal or gastric varices. Initial attempts at splenic embolization resulted in a multitude of serious complications, including splenic abscess, infarction, cysts, and contrast-induced renal insufficiency (12). More recently, better results have been obtained by means of modified techniques aimed at PSE. In this study, we reported technical success of 92.6%, as there were seven (7.4%) unsuccessful cases of the 95 studied patients. In six patients, the procedure failed because of failure of catheterization of the distal segment of the splenic artery, the cobra catheter (5-Fr) could not be introduced into the distal segment of the splenic artery as it was greatly hypertrophied and highly tortuous. In one patient, the procedure failed because total rather than PSE was accidentally performed due to the use of bigger gel foam cubes. Our technical success rates are comparable with those reported by Hagiwara et al. In their study, they performed splenic embolization for management of splenic trauma in 15 patients (13). They reported success rate of 86.6% as the technique failed in 2 of 15 studied patients. On the Other hand; He et al. (14), performed PSE for 34 patients with hypersplenism and reported 100% technical success. In this study, none of our patients developed any life-threatening complications such as splenic rupture, splenic abscess, pancreatitis, portal, or splenic vein thrombosis. Post-embolization syndrome (pain, fever, and vomiting) was the most common adverse event, followed by left sided pleural effusion in about one-third of patients. These findings are in agreement with those of (15) who studied the side effects of PSE, encountered in 17 cirrhotic patients with hypersplenism and portal hypertension (15). The most common side effects were abdominal pain (82.4%) and fever (94.1%). However, they reported serious side effects in two patients; one of them developed acute on top of chronic liver failure and the other developed splenic abscess.

© 2015 APMIS. Published by John Wiley & Sons Ltd

Splenic span increased by 3, 5, and 8% after 3, 7, and 30 days of PSE, respectively, this increase could be attributed to necrosis and edema of the surrounding splenic parenchyma with subsequent distension of the splenic capsule. This increase is in agreement with (16), who performed the procedure on five cirrhotic children for the treatment of hypersplenism (16). They showed an increase in 10– 40% in the splenic volume using dynamic CT scan 2 weeks after PSE. They showed a higher risk of splenic rupture within the first 4 weeks after PSE as the splenic volume markedly increased. Accordingly, patients should be instructed to avoid any blunt trauma to the upper abdomen for approximately 1 month after PSE. The increase in splenic size in our patients was less than that of Watanabe et al. possibly because of different age groups and different methodology for estimating splenic size. Previous studies addressing the hemodynamic effects of PSE on portal and mesenteric circulations were contradictory. Authors described different methods for hemodynamic assessment. In 1987, Nishida and his group studied partial occlusion of the splenic artery, using a balloon catheter (17). They recorded the direct portal pressure using the trans-hepatic approach and measured the flow volumes of the splenic, portal, and superior mesenteric veins using duplex ultrasonography. They found that splenic arterial occlusion caused marked portal decompression, significant reduction in splenic venous flow with minimal decrease in portal vein flow volume. On the other hand, the superior mesenteric venous flow, unexpectedly, showed marked increase after occlusion. Accordingly, Nishida et al. concluded that there was a reciprocal relationship between the splenic and superior mesenteric venous flow following occlusion of the splenic artery; a decrease in flow in the former was associated with an increase in flow in the latter that maintained average portal vein flow volume. On the other hand, Gusberg et al. found no significant portal decompression after clamping the proximal splenic vein during distal spleno-renal shunt operations to create PSE-like state (18). Although they did not measure the blood flow of the portal vein, superior mesenteric vein, or superior mesenteric artery, they suggested that the pressure was not affected due to compensatory increased flow in the portal vein, mostly from the secondary increased flow of the superior mesenteric flow. Mukaiya et al. studied the portal hemodynamics in seven cirrhotic patients after PSE (19). They measured direct venous pressure and flow volume of the portal and splenic veins using percutaneous portography. The changes in portal blood flow/

1037

HELALY et al.

pressure and hemodynamics were examined by a thermodilution method. PSE reduced the direct portal pressure from 48 to 42 cm H2O. HPS induced a decrease in the blood flow of the splenic artery and in the splenic vein pressure without decreasing the portal blood flow. In this study, duplex indices reflecting the portal hypertensive state showed significant portal decompression after PSE. The mean values of CI which reflects the degree of portal vein congestion and compliance, significantly decreased from 0.12 to 0.06 cm s 1 week post-PSE. This decrease was due to the increase in portal vein velocity as well as the decrease in CSA. Moreover, we found that the hepatic arterial impedance indices, the pulsatility and resistive indices significantly decreased compared to pre-embolization values. This decrease indicates a decrease in the sinusoidal pressure that reflects portal hypertension. The LVI, significantly increased in a chronological manner after PSE. The decrease in hepatic artery compliance indices (RI and PI) and increase in LVI, collectively, reflect improvement of the portal venous and hepatic arterial hypertensive state which is in agreement with the studies of Nishida et al. (17) and Mukayia et al. (19). Regarding the blood flow volume patterns within the portal circulation, we did not record a significant difference in portal vein flow volume after PSE. These results agreed with those recorded in previous studies (17–19). Regarding the mesenteric artery flow, we did not find any significant difference after embolization. These results disagreed with those reported by Nishida et al. (17), Mukayia et al. (19) and Gusberg et al. (18). Thus, the assumption of reciprocal inter-relationship between the splenic and superior mesenteric venous flow, reported by the previous authors, needs to be revised in future studies on a larger number of patients. Maintenance of the portal venous flow constant depends mainly on the status of portal arterial inflow and the presence of a collateral circulation. In the absence of collateral flow, the portal flow is maintained by the increased superior mesenteric venous flow (17). In the presence of collateral circulation, the portal vein resumes its constant flow by a distinctive way, as postulated by Nakano et al. (20), both the splenic and superior mesenteric arteries have significantly high flow volumes in case of portal hypertension, especially in the presence of esophageal varices. This increased flow volume in both arteries, collectively, is responsible not only for the maintenance of normal portal venous flow volume but also for the development of collateral circula-

1038

tion. Consequently, it is responsible also for the retrograde (hepatofugal) flow in the left and short gastric veins, supplying the esophageal and gastric varices. After splenic artery embolization, based on 60–80% dearterialization in our technique, the splenic venous outflow decreased in a parallel pattern. The pressure also decreased, and consequently the left and short gastric veins resumed their normal antegrade flow (hepatopetal direction) with subsequent collapse of the varices. Thus, the remaining 20–40% of the splenic venous blood together with the resumed antegrade flow of left and short gastric veins plus the superior mesenteric venous flow made the portal vein flow constant. Taniai et al. reported that occlusion of the splenic arterial flow by PSE blocked indirectly the blood flow to the short and left gastric veins (21). Although they followed the technique of 50% dearterialization of the spleen, they noted, at the post-embolization films, the absence of the retrograde flow in these veins. Our finding of significant portal decompression is also supported by many endoscopic, clinical and scintigraphic studies. PSE was reported to be beneficial in reducing risk of variceal bleeding. Koconis et al. (22) reported that PSE decreased the annual incidence of variceal hemorrhage by 80%. Palsson et al. (23) performed PSE in patients with history of bleeding esophageal varices and thrombocytopenia which resulted in the reduction of the number of variceal bleeding episodes from a mean of 4.3  2.9 prior to treatment to 1.1  1.3 after PSE (p < 0.001). Ohmoto et al. (24) conducted a study on cirrhotic patients with large esophageal varices and thrombocytopenia. Patients were treated with either endoscopic variceal ligation (EVL) alone or combination therapy of EVL and PSE. The combination therapy cohort demonstrated a significant reduction in the development of new varices from 88 to 67% (p = 0.038), decreased episodes of variceal bleeding from 34 to 17% (p = 0.024), and improved overall survival from 31 to 50% (p = 0.042). Our results, regarding portal decompression, are not in agreement with those reported by Gusberg et al. (18). We believe that, the main cause of discrepancy is related to the technique used by Gusberg et al. (18) as they ligated the splenic vein surgically and measured the direct portal pressure intra-operatively, which might be affected by anesthesia, blood loss, and duration of the operation, unlike PSE performed percutaneously which is quite safer. In conclusion, PSE had dual benefits regarding the hematological and hemodynamic outcome. It improved the hematological profile as well as

© 2015 APMIS. Published by John Wiley & Sons Ltd

PORTAL HEMODYNAMICS AFTER PSE

DISCUSSION Bleeding from esophageal varices is a common and life-threatening complication in patients with cirrhosis and portal hypertension. Several approaches to control bleeding from esophageal varices were developed to lower either the resistance (i.e., shunt surgery) or the portal inflow (i.e., beta adrenergic blockers or PSE). The effectiveness of these approaches on portal decompression would be dependent, at least in part, on whether increased resistance or increased inflow is the predominant mechanism for portal hypertension (11). Improving portal hemodynamics is usuaslly associated with a reduction in the sequelae of portal hypertension, particularly, bleeding esophageal or gastric varices. Initial attempts at splenic embolization resulted in a multitude of serious complications, including splenic abscess, infarction, cysts, and contrast-induced renal insufficiency (12). More recently, better results have been obtained by means of modified techniques aimed at PSE. In this study, we reported technical success of 92.6%, as there were seven (7.4%) unsuccessful cases of the 95 studied patients. In six patients, the procedure failed because of failure of catheterization of the distal segment of the splenic artery, the cobra catheter (5-Fr) could not be introduced into the distal segment of the splenic artery as it was greatly hypertrophied and highly tortuous. In one patient, the procedure failed because total rather than PSE was accidentally performed due to the use of bigger gel foam cubes. Our technical success rates are comparable with those reported by Hagiwara et al. In their study, they performed splenic embolization for management of splenic trauma in 15 patients (13). They reported success rate of 86.6% as the technique failed in 2 of 15 studied patients. On the Other hand; He et al. (14), performed PSE for 34 patients with hypersplenism and reported 100% technical success. In this study, none of our patients developed any life-threatening complications such as splenic rupture, splenic abscess, pancreatitis, portal, or splenic vein thrombosis. Post-embolization syndrome (pain, fever, and vomiting) was the most common adverse event, followed by left sided pleural effusion in about one-third of patients. These findings are in agreement with those of (15) who studied the side effects of PSE, encountered in 17 cirrhotic patients with hypersplenism and portal hypertension (15). The most common side effects were abdominal pain (82.4%) and fever (94.1%). However, they reported serious side effects in two patients; one of them developed acute on top of chronic liver failure and the other developed splenic abscess.

© 2015 APMIS. Published by John Wiley & Sons Ltd

Splenic span increased by 3, 5, and 8% after 3, 7, and 30 days of PSE, respectively, this increase could be attributed to necrosis and edema of the surrounding splenic parenchyma with subsequent distension of the splenic capsule. This increase is in agreement with (16), who performed the procedure on five cirrhotic children for the treatment of hypersplenism (16). They showed an increase in 10– 40% in the splenic volume using dynamic CT scan 2 weeks after PSE. They showed a higher risk of splenic rupture within the first 4 weeks after PSE as the splenic volume markedly increased. Accordingly, patients should be instructed to avoid any blunt trauma to the upper abdomen for approximately 1 month after PSE. The increase in splenic size in our patients was less than that of Watanabe et al. possibly because of different age groups and different methodology for estimating splenic size. Previous studies addressing the hemodynamic effects of PSE on portal and mesenteric circulations were contradictory. Authors described different methods for hemodynamic assessment. In 1987, Nishida and his group studied partial occlusion of the splenic artery, using a balloon catheter (17). They recorded the direct portal pressure using the trans-hepatic approach and measured the flow volumes of the splenic, portal, and superior mesenteric veins using duplex ultrasonography. They found that splenic arterial occlusion caused marked portal decompression, significant reduction in splenic venous flow with minimal decrease in portal vein flow volume. On the other hand, the superior mesenteric venous flow, unexpectedly, showed marked increase after occlusion. Accordingly, Nishida et al. concluded that there was a reciprocal relationship between the splenic and superior mesenteric venous flow following occlusion of the splenic artery; a decrease in flow in the former was associated with an increase in flow in the latter that maintained average portal vein flow volume. On the other hand, Gusberg et al. found no significant portal decompression after clamping the proximal splenic vein during distal spleno-renal shunt operations to create PSE-like state (18). Although they did not measure the blood flow of the portal vein, superior mesenteric vein, or superior mesenteric artery, they suggested that the pressure was not affected due to compensatory increased flow in the portal vein, mostly from the secondary increased flow of the superior mesenteric flow. Mukaiya et al. studied the portal hemodynamics in seven cirrhotic patients after PSE (19). They measured direct venous pressure and flow volume of the portal and splenic veins using percutaneous portography. The changes in portal blood flow/

1037