JOURNAL
OF SURGICAL
RESEARCH
53,621-624
(19%)
Effects of Splenectomy and Splenic Artery Ligation on the Portal Pressure in Portal Hypertensive Rats PIN-WEN Department
of Surgery,
Medical
LIN, M.D.,
College, National
AND YEN-SHENG SHAN, M.D.
Cheng Kung University,
138 Sheng-Li
Road, Tainan,
Taiwan,
Republic of China
Submitted for publication August 13, 1991
Immediate, short-, and long-term effects of splenectomy and splenic artery ligation on the portal pressure were studied in animal models experimentally created by partial portal vein ligation. The portal pressure of these animals would usually elevate immediately after partial ligation of the portal vein from a normal level of 6.0 + 0.5 to 14.8 -+ 1.3 mm Hg (P < 0.005), which could be maintained at least for 6 months. The portal pressure measured at 2 weeks, 4 weeks, and 6 months after portal vein ligation was 14.0 + 2.7, 15.2 & 2.7, and 12.7 + 2.0 mm Hg, respectively (P < 0.005, as compared with the normal). When splenectomy was performed on these animals at 2 weeks after partial portal vein ligation, the pressure dropped immediately but only transiently from 14.0 f 2.7 to 11.0 + 3.0 mm Hg, and bounced back to the presplenectomy level in 20 sec. After an additional 2 weeks, the portal pressure in these splenectomized rats was usually at 15.2 f 4.2 mm Hg, which was indistinguishable from that of rats whose portal vein was ligated but the spleen was not removed. Six months after splenectomy, however, the portal hypertensive rats had a portal pressure of 17.1+ 6.4 mm Hg, which was significantly higher than that of the controls. Splenic artery ligation, on the other hand, did not result in any immediate decrease in portal pressure (14.0 f 2.7 mm Hg vs 14.6 f 1.4 mm Hg; P > 0.1). However, at 2 weeks, animals that had undergone splenic artery ligation had significantly lower portal pressure than the control animals (12.0 f 1.9 mm Hg vs 15.4 * 2.7 mm Hg; P < 0.05), and the difference in the pressure between these two groups became even more pronounced at 6 months (10.5 t 2.6 mm Hg vs 15.4 k 2.7 mm Hg; P < 0.05). We concluded that in the longrun splenectomy would not lower the portal pressure of portal hypertensive rats, while ligation of the splenic 0 1992 Academic Press, Inc. artery would.
INTRODUCTION
Portal hypertension is one of the most serious complications of liver cirrhosis and the single most common clinical manifestation of portal hypertension, regardless
of its cause, is splenomegaly [ 1,2]. Therefore, splenomegaly has long been considered an active element of the disease process [3]. Such a concept led to the inclusion of splenectomy as one of the most effective therapeutic modalities for the disease [4-61. Yet, splenectomy in portal hypertensive patients has met only with varying degrees of success and in fact splenectomy alone in this condition is considered an ineffective therapy. These observations led us to believe that the spleen may play a regulatory function, as a reservoir for portal blood in portal hypertension, and its removal may not be as beneficial to the patient as is generally believed. Obviously, studies with animal models are required in order to shed light on this intriguing problem. In order to investigate the significance of the spleen in portal hypertension, we attempted to evaluate the effect of its removal or ligation of the splenic artery on the pressure of the portal system in the animal model. The immediate, short-, and long-term effects of these surgical procedures in these animals were determined and are reported herein. MATERIALS
AND
METHODS
Animals Two-month-old male Wistar rats weighing 260-300 g were housed in stainless-steel, screen-floored cages. The animals were allowed free access to rat chow and water until the time of experiment. Creation of Portal Hypertension Ligation (PVL)
by Partial
Portal
Vein
Each animal was anesthetized with an intraperitoneal injection of ketamine at 10 g/100 g body weight. PVL was performed as described by Groszmann et al. [ 7-101. Briefly, the abdomen was opened through a midline incision. The omentum and part of the intestine were gently lifted from the abdominal cavity and kept moist with warm, normal-saline-moistened gauze. The portal vein was exposed and carefully isolated free from the hepatic artery and the bile duct. A 20-gauge, blunt-end 621
oozz-4804/92
Copyright Q 1992 All
rights
of reproduction
by
$4.00 Inc. reserved.
AcademicPress, in any
form
622
JOURNAL
OF SURGICAL
RESEARCH:
needle was placed alongside the length of the portal vein, and one ligature of 000 silk placed proximal to the bifurcation of the vein was tied snugly. The needle was then removed, and the portal vein was allowed to reexpand distal to the stenosis. The abdominal viscera were replaced into the abdomen, which was closed in layers with sutures. Measurement
of the Portal Pressure
The abdomen was opened through a midline incision. The portal vein was exposed, and a 21-gauge butterfly needle was inserted into the portal vein. Portal pressure was measured with a transducer connected to the butterfly needle and was recorded continuously on a polygraph. The external zero reference point was placed at the midportion of the animal. A bolus of saturated KC1 was used to sacrifice the animal at the completion of the measurement of portal pressure. Splenectomy The abdomen was opened through a midline incision. The spleen was mobilized to the center of the operation field after dissecting the surrounding ligaments. Two ligatures of 000 silk were utilized to ligate the hilar vessels. The spleen was then removed and weighed. Splenic Artery Ligation
(SAL)
The abdomen was opened through a midline incision. The spleen was mobilized to the center of the operation field after dissecting the surrounding ligaments. One ligature of 000 silk was utilized to ligate the splenic artery close to the hilum. Then, the spleen was placed back to its bed. Grouping of Animals
and the Experimental
Design
Two-month-old rats, 108 total number, were divided into three groups of 36 animals each, designated as groups A, B, C, respectively. All animals in these three groups were subject to PVL to create portal hypertension. An additional group of 36 rats was designated group D, which was not subject to operation and served as normal controls. Two weeks after the PVL, animals in group A were subject to splenectomy and those in group B to SAL. Only PVL was performed in group C. The portal pressure in 12 animals from each of the four groups was measured at three time points, namely, immediately, and 2 weeks and 6 months after the operation was performed. For the convenience of identification, a system of abbreviations is used in this report for the animals which had undergone different operations at different time points. For example, a rat which had undergone PVL and whose portal pressure was measured immediately thereafter is designated PVL-imm, while a rat which had in addition undergone splenectomy and whose portal pressure was
VOL.
53, NO. 6, DECEMBER
1992
measured at 2 weeks is designated spl-2W. A rat which had undergone SAL and whose portal pressure was measured at 6 months is designated SAL-GM and so on. Data Analysis The results are expressed as means f SD. Statistical analysis using the nonparameter Mann-Whitney test was performed between groups. Results were considered significant at P < 0.05. When 0.05 < P < 0.1, a forward significance was considered. RESULTS
The baseline data of portal pressure and the weight of the spleen were obtained from group D as shown in Table 1. A part of the results was published previously [ 111. Effects of splenectomy and SAL on the portal pressure in the portal hypertensive rats are shown in Fig. 1. Immediately upon partial ligation of the portal vein, the portal pressure rose from 6.0 f 0.5 to 14.8 f 1.3 mm Hg (P < 0.005). The portal pressure measured at 2 weeks, 4 weeks, and 6 months after PVL was 14.0 f 2.7, 15.2 + 2.7, and 12.7 + 2.0 mm Hg, respectively (P < 0.005, as compared with normal controls). The portal pressure in PVL-GM rat was significantly lower than in the PVL2W and PVL-4W rats. When splenectomy was performed in the PVL-2W rat, the portal pressure immediately dropped from 14.1 & 2.7 to 11.0 + 3.1 mm Hg (P < O.Ol), but returned to the presplenectomy level within 20 sec. It was noteworthy that although it dropped significantly, the pressure never reached the pre-PVL level, and in fact was maintained at a significantly higher level than that of the normal (P < 0.005). The portal pressure of the spl-2W rats was similar to that of the PVL-2W and PVL-4W rats but higher than that of N-4W rats (15.2 + 4.2,14.0 + 2.7, and 15.2 f 2.7 vs 7.8 + 1.1 mm Hg; P < 0.005). The portal pressure of the spl-GM rats was significantly higher than that of the PVL-GM rats (17.1 * 6.4 vs 12.7 + 2.0 mm Hg; P < 0.05). Though SAL did not show any immediate decrease in portal pressure (14.0 f 2.7 mm Hg vs 14.6 -+ 1.4 mm Hg; P > 0.05), a significant decrease of portal pressure (15.2 f 2.7 mm Hg vs 12.0 + 1.9 mm Hg; P < 0.05) was noted in the SAL-2W rats. The decrease in portal pressure was more obvious (12.7 ? 2.0 vs 10.5 + 2.6 mm Hg; P < 0.05) in SAL-GM rats. The portal pressure in these rats was statistically lower than in PVL-GM rats (P < 0.05). Significant splenomegaly was noted in the PVL and SAL rats as compared with the normal controls, but the weight of the spleen in the SAL rats was lower than in the PVL rats (P -C 0.05). DISCUSSION
A sustained portal hypertension could be r ;producibly induced by PVL in rats [lo]. The portal pressure rose
LIN
Results
BW LW LW/BW (x10-2) SW SW/SW (x10-3)
of Splenectomy
SHAN:
PORTAL
and Splenic
HYPERTENSION
TABLE
1
Arterial
Ligation
623
IN RATS
on Portal
N-6M (n = 12)
PVL-PW (n = 12)
PVL-4w (n = 12)
PVL-GM (n = 12)
270 f 5 10.60 f 1.15
351 f 35 11.17 f 0.82
525 + 48 16.04 + 2.45
315 + 16 9.41 -t 1.61
356 + 39 9.64 + 1.96
542 -+ 24 13.59 t 0.64
352 + 29 9.55 * 1.2
449 k 44 10.06 t 1.95
324 + 43 12.21 + 1.59
493 k 57 11.26 + 1.80
3.84 0.97 + 0.09
3.18 0.95 + 0.07
3.06 0.95 i- 0.08
2.99 1.76 t 1.61
2.71 1.33 i 0.22
2.59 1.69 t 0.16
2.71 1.88 + 0.60
2.01 1.83 i 0.34
3.55 1.39 2 0.36
2.99 1.24 t 0.23
3.51
2.70
1.80
5.59 14.0 i 2.7 11.0 ?I 3.0* 14.6 k 1.4#
3.71
3.22
5.98
6.13
4.05
2.52
15.2 T 2.7
12.7 _t 2.0**
15.2 i 4.2@
17.1 + 6.4***
12.0 + 1.9##
10.5 f 2.6###
7.7 AI 0.8
Spl-GM (n = 12)
Rats
N-4W (n = 12)
7.8 + 1.1
Spl-PW (n = 12)
Hypertensive
N-l (n = 12)
6.0 f 0.5
PP
AND
SAL-SW (n = 12)
SAL-GM (n = 12)
Note. BW: body weight, g. LW: liver weight, g. SW: spleen weight, g. PP: portal pressure, mm Hg. N-l: normal rats of 2 months old. N-4W: N-l rats kept for 4 weeks more. PVL-PW: 2 weeks after portal vein ligation. Spl-2W: 2 weeks after splenectomy on the PVL-2W rats. SAL-PW: 2 weeks after splenic arterial ligation on the PVL-2W rats. * Spl-imm, P < 0.01 compared with PVL-2W. # SAL-imm, I’ > 0.1 compared with PVL-2W. ** P < 0.05 compared with N-6M; P < 0.01 compared with PVL-SW. @P > 0.1 compared with PVL-4W. *** P -C 0.05 compared with PVL-GM; P < 0.005 compared with N-6M. ## P < 0.005 compared with N-4W; P c 0.05 compared with PVL-4W. ### P < 0.05 compared with PVL-GM; P < 0.005 compared with Spl-GM, P < 0.05 compared with N-6M.
immediately after PVL and had been maintained thereafter, Sikuler and Groszmann [lo] stated that the mechanisms which contributed to the development and maintenance of portal hypertension change with time. The immediate rise in portal pressure after PVL was attributable to the abrupt increase in portal venous resistance. In addition to increased portal venous resistance, a hyperdynamic circulatory derangement with splanchnit hyperemia (or increased portal venous inflow) per-
20.00. 3 I
H
normal
-
PVL-only
w
PVL-spl
0-0
PVL-SAL
rat*
5 0 15.00
10.00
5.00
0.00
+ 0
2w
4W
6M TIME
FIG. 1. Effects of splenectomy and splenic artery ligation porta pressure in portal hypertensive rats.
on the
haps played an even more important role in the maintenance of short-term portal hypertension. This hyperdynamic circulation, however, was no longer present in later stages of portal hypertension. The portal pressure in this stage was primarily maintained by an increased resistance to the portal venous inflow. In our study, portal pressure of PVL-GM rats was significantly lower than that of PVL-2W rats, while it was still much higher than that of the normal controls. However, in their report [lo], Sikuler and Groszmann noted that the pressure in the long-term portal hypertensive rats was lower, but not statistically significant, than in the short-term group. There are still no solid experimental data to satisfactorily explain this phenomenon. However, it can be speculated that the force of hemodynamic homeostasis may have led the pressure to settle at a new, lower, and physiologically acceptable level in the long-term portal hypertensive rats. Regardless of what the pathophysiology of portal hypertension is [12-151, in this study we have observed that SAL was effective in reducing the portal pressure following PVL, while splenectomy was ineffective. This was most clearly demonstrated by the fact that splenectomy only led to a 20-set drop in the portal pressure, which bounced back and became even higher than the presplenectomy level. The drop obviously was due to a very transient cut-off in splenic flow, which was more than adequately compensated by an increase in the mesenteric blood flow [ 161. The spleen, moreover, may act as an effective circulatory reservoir with its potential to develop extensive collateral pathways [ 171. In addition, it is possible that the spleen may be capable of producing some still unidentified humoral factors which could regu-
624
JOURNAL
OF SURGICAL
RESEARCH:
late the portal pressure. Removal of the spleen will effectively eliminate these advantages from the body and leave the portal pressure unharnessed. It is interesting to note that our experimental data could be invoked to explain the observations made by Koyama et al. in a clinical setting [16]. They observed that the portal pressure in patients, who had undergone splenectomy for idiopathic portal hypertension several years before, was distinctly increased. They speculated that this was due to increased superior mesenteric blood flow and/or increased block in the liver in these patients. When SAL was performed in the PVL-2W rat, no significant decrease in portal pressure was observed immediately but was 2 weeks after the procedure. A further drop in the portal pressure was observed in the ensuing 6 months. Three explanations could be offered for this observation. One is that SAL cut off the splenic arterial inflow, which has known to be increased in portal hypertension [7, 81. In their report, Witte et al. noted that ligation of the splenic artery close to the splenic hilum would interrupt the splenic inflow and promoted splenic infarction and thereby reduction in functional splenic mass [ 17,181. This phenomenon was also clearly shown in our study; the splenic weight of the splenic artery ligated group was statistically lower than that of the unligated group. Second, with the ligated spleen in situ, the porto-systemic shunts in and around the spleen were left intact, which could lower the portal pressure. Moreover, these porto-systemic venous channels no longer draining a hyperdynamic splenic circuit, which was interrupted by splenic artery ligation, may be converted into an escape route for mesenteric venous blood entering the portal system under higher pressure [17, 181. Third, since in SAL rats, the spleen was not removed, its above-mentioned regulatory functions on portal pressure were still at work. Therefore, the pressure determined in these animals was significantly lower than that of splenectomized rats. The spleen, therefore, appears to have a modulating function on the portal pressure in this animal model, and ligation of splenic artery may bring about a better effect for portal hypertensive rats than splenectomy. We concluded that in the long run splenectomy would not lower the portal pressure of portal hypertensive rats, while ligation of the splenic artery would.
ACKNOWLEDGMENTS The authors thank H. Y. Chang for statistical support, H. H. Hou for typing the manuscript, and Prof K. Y. Huang for reviewing and revising the manuscript.
VOL.
53, NO. 6, DECEMBER
1992
REFERENCES 1.
Dumont, A. E., Amorosi, E., and Stahl, W. M. Significance of splenomegaly in patients with hepatic cirrhosis and bleeding esophageal varices. Ann. Surg. 171: 522, 1971. 2. Dumont, A. E., Berman, I. R., Stahl, W. M., and Ring, S. M. Significance of an enlarged splenic artery in patients with bleeding varices. Ann. Surg. 175: 466, 1972. 3. Witte, C. L., Witte, M. H., Bair, G., Mobley, W. P., and Morton, D. Experimental study of hyperdynamic vs. stagnant mesenteric blood flow in portal hypertension. Ann. Surg. 179: 304, 1974. 4. Hassab, M. A. Gastroesophageal decongestion and splenectomy in the treatment of esophageal varices in bilharzial cirrhosis: Further studies with a report on 355 operations. Surgery 61: 169, 1967. 5. Sugiura, M., and Futagawa, S. A new technique for treating esophageal varices. J. Thorac. Cardiouasc. Surg. 66: 677, 1973. 6. Sugiura, M., and Futagawa, S. Esophageal transection with paraesophagogastric devascularizations (the Sugiura Procedure) in the treatment of esophageal varices. World J. Surg. 8: 673,1984. 7. Vorobioff, J., Bredfeldt, J. E., and Groszmann, R. J. Hyperdynamic circulation in portal-hypertensive rat model: A primary factor for maintenance of chronic portal hypertension. Am. J. Physiol. 144: G52, 1983. 8. Groszmann, R. J., Vorobioff, J., and Riley, E. Splanchnic hemodynamics in portal hypertensive rats: Measurement with r-labeled microspheres. Am. J. Physiol. 242: G156,1983. 9. Sikuler, E., Kravetz, D., and Groszmann, R. J. Evolution of portal hypertension and mechanisms involved in its maintenance in a rat model. Am. J. Physiol. 248: G618,1985. 10. Sikuler, E., and Groszmann, R. J. Hemodynamic studies in longand short-term portal hypertensive rats: The relation to systemic glucagon level. Hepatology 6: 414, 1986. 11. Lin, P. W., and Shan, Y. S. Effects of splenectomy on portal pressure in portal hypertensive rats. J. Surg. Assoc. ROC 23: 208, 1990. 12. Benoit, J. N., and Granger, D. N. Splanchnic hemodynamics in chronic portal hypertension. Semin. Liver Dis. 6: 287, 1987. 13. Benoit, J. N., Womack, W. A., Hernandez, L., and Granger, D. N. “Forward” and “backward” flow mechanisms of portal hypertension, relative contributions in the rat model of portal vein stenosis. Gastroenterol. 89: 1092, 1985. J. A., Harper, S. L., Kvietys, P. R., 14. Benoit, J. N., Barrowman, and Granger, D. N. Role of humoral factors in the intestinal hyperemia associated with chronic portal hypertension. Am. J. Physiol. 247: G486, 1984. of flow and resis15. Sikuler, E., and Groszmann, R. J. Interaction tance in maintenance of portal hypertension in a rat model. Am. J. Physiol. 250: G205, 1986. 16. Koyama, K., Ouchi, K., and Sato, T. Development mechanism of portal hypertension in idiopathic portal hypertension. In K. Okuda, and M. Omata (Eds.), Idiopathic Portal Hypertension. Univ. of Tokyo Press, 1983. Pp. 231-240. 17. Witte, C. L., Witte, M. H., Renert, W., O’Mara, R. E., and Lilien, D. L. Splenic artery ligation in selected patients with hepatic cirrhosis and in Sprague-Dawley rats. Surg. Gynecol. Obstet. 142: 1,1976. 18. Witte, C. L., Corrigan, J. J., Witte, M. H., and O’Mara, R. E. Splenic artery ligation in experimental hypersplenism. Surg. 80: 581,1976.
OF SURGICAL
RESEARCH
53,621-624
(19%)
Effects of Splenectomy and Splenic Artery Ligation on the Portal Pressure in Portal Hypertensive Rats PIN-WEN Department
of Surgery,
Medical
LIN, M.D.,
College, National
AND YEN-SHENG SHAN, M.D.
Cheng Kung University,
138 Sheng-Li
Road, Tainan,
Taiwan,
Republic of China
Submitted for publication August 13, 1991
Immediate, short-, and long-term effects of splenectomy and splenic artery ligation on the portal pressure were studied in animal models experimentally created by partial portal vein ligation. The portal pressure of these animals would usually elevate immediately after partial ligation of the portal vein from a normal level of 6.0 + 0.5 to 14.8 -+ 1.3 mm Hg (P < 0.005), which could be maintained at least for 6 months. The portal pressure measured at 2 weeks, 4 weeks, and 6 months after portal vein ligation was 14.0 + 2.7, 15.2 & 2.7, and 12.7 + 2.0 mm Hg, respectively (P < 0.005, as compared with the normal). When splenectomy was performed on these animals at 2 weeks after partial portal vein ligation, the pressure dropped immediately but only transiently from 14.0 f 2.7 to 11.0 + 3.0 mm Hg, and bounced back to the presplenectomy level in 20 sec. After an additional 2 weeks, the portal pressure in these splenectomized rats was usually at 15.2 f 4.2 mm Hg, which was indistinguishable from that of rats whose portal vein was ligated but the spleen was not removed. Six months after splenectomy, however, the portal hypertensive rats had a portal pressure of 17.1+ 6.4 mm Hg, which was significantly higher than that of the controls. Splenic artery ligation, on the other hand, did not result in any immediate decrease in portal pressure (14.0 f 2.7 mm Hg vs 14.6 f 1.4 mm Hg; P > 0.1). However, at 2 weeks, animals that had undergone splenic artery ligation had significantly lower portal pressure than the control animals (12.0 f 1.9 mm Hg vs 15.4 * 2.7 mm Hg; P < 0.05), and the difference in the pressure between these two groups became even more pronounced at 6 months (10.5 t 2.6 mm Hg vs 15.4 k 2.7 mm Hg; P < 0.05). We concluded that in the longrun splenectomy would not lower the portal pressure of portal hypertensive rats, while ligation of the splenic 0 1992 Academic Press, Inc. artery would.
INTRODUCTION
Portal hypertension is one of the most serious complications of liver cirrhosis and the single most common clinical manifestation of portal hypertension, regardless
of its cause, is splenomegaly [ 1,2]. Therefore, splenomegaly has long been considered an active element of the disease process [3]. Such a concept led to the inclusion of splenectomy as one of the most effective therapeutic modalities for the disease [4-61. Yet, splenectomy in portal hypertensive patients has met only with varying degrees of success and in fact splenectomy alone in this condition is considered an ineffective therapy. These observations led us to believe that the spleen may play a regulatory function, as a reservoir for portal blood in portal hypertension, and its removal may not be as beneficial to the patient as is generally believed. Obviously, studies with animal models are required in order to shed light on this intriguing problem. In order to investigate the significance of the spleen in portal hypertension, we attempted to evaluate the effect of its removal or ligation of the splenic artery on the pressure of the portal system in the animal model. The immediate, short-, and long-term effects of these surgical procedures in these animals were determined and are reported herein. MATERIALS
AND
METHODS
Animals Two-month-old male Wistar rats weighing 260-300 g were housed in stainless-steel, screen-floored cages. The animals were allowed free access to rat chow and water until the time of experiment. Creation of Portal Hypertension Ligation (PVL)
by Partial
Portal
Vein
Each animal was anesthetized with an intraperitoneal injection of ketamine at 10 g/100 g body weight. PVL was performed as described by Groszmann et al. [ 7-101. Briefly, the abdomen was opened through a midline incision. The omentum and part of the intestine were gently lifted from the abdominal cavity and kept moist with warm, normal-saline-moistened gauze. The portal vein was exposed and carefully isolated free from the hepatic artery and the bile duct. A 20-gauge, blunt-end 621
oozz-4804/92
Copyright Q 1992 All
rights
of reproduction
by
$4.00 Inc. reserved.
AcademicPress, in any
form
622
JOURNAL
OF SURGICAL
RESEARCH:
needle was placed alongside the length of the portal vein, and one ligature of 000 silk placed proximal to the bifurcation of the vein was tied snugly. The needle was then removed, and the portal vein was allowed to reexpand distal to the stenosis. The abdominal viscera were replaced into the abdomen, which was closed in layers with sutures. Measurement
of the Portal Pressure
The abdomen was opened through a midline incision. The portal vein was exposed, and a 21-gauge butterfly needle was inserted into the portal vein. Portal pressure was measured with a transducer connected to the butterfly needle and was recorded continuously on a polygraph. The external zero reference point was placed at the midportion of the animal. A bolus of saturated KC1 was used to sacrifice the animal at the completion of the measurement of portal pressure. Splenectomy The abdomen was opened through a midline incision. The spleen was mobilized to the center of the operation field after dissecting the surrounding ligaments. Two ligatures of 000 silk were utilized to ligate the hilar vessels. The spleen was then removed and weighed. Splenic Artery Ligation
(SAL)
The abdomen was opened through a midline incision. The spleen was mobilized to the center of the operation field after dissecting the surrounding ligaments. One ligature of 000 silk was utilized to ligate the splenic artery close to the hilum. Then, the spleen was placed back to its bed. Grouping of Animals
and the Experimental
Design
Two-month-old rats, 108 total number, were divided into three groups of 36 animals each, designated as groups A, B, C, respectively. All animals in these three groups were subject to PVL to create portal hypertension. An additional group of 36 rats was designated group D, which was not subject to operation and served as normal controls. Two weeks after the PVL, animals in group A were subject to splenectomy and those in group B to SAL. Only PVL was performed in group C. The portal pressure in 12 animals from each of the four groups was measured at three time points, namely, immediately, and 2 weeks and 6 months after the operation was performed. For the convenience of identification, a system of abbreviations is used in this report for the animals which had undergone different operations at different time points. For example, a rat which had undergone PVL and whose portal pressure was measured immediately thereafter is designated PVL-imm, while a rat which had in addition undergone splenectomy and whose portal pressure was
VOL.
53, NO. 6, DECEMBER
1992
measured at 2 weeks is designated spl-2W. A rat which had undergone SAL and whose portal pressure was measured at 6 months is designated SAL-GM and so on. Data Analysis The results are expressed as means f SD. Statistical analysis using the nonparameter Mann-Whitney test was performed between groups. Results were considered significant at P < 0.05. When 0.05 < P < 0.1, a forward significance was considered. RESULTS
The baseline data of portal pressure and the weight of the spleen were obtained from group D as shown in Table 1. A part of the results was published previously [ 111. Effects of splenectomy and SAL on the portal pressure in the portal hypertensive rats are shown in Fig. 1. Immediately upon partial ligation of the portal vein, the portal pressure rose from 6.0 f 0.5 to 14.8 f 1.3 mm Hg (P < 0.005). The portal pressure measured at 2 weeks, 4 weeks, and 6 months after PVL was 14.0 f 2.7, 15.2 + 2.7, and 12.7 + 2.0 mm Hg, respectively (P < 0.005, as compared with normal controls). The portal pressure in PVL-GM rat was significantly lower than in the PVL2W and PVL-4W rats. When splenectomy was performed in the PVL-2W rat, the portal pressure immediately dropped from 14.1 & 2.7 to 11.0 + 3.1 mm Hg (P < O.Ol), but returned to the presplenectomy level within 20 sec. It was noteworthy that although it dropped significantly, the pressure never reached the pre-PVL level, and in fact was maintained at a significantly higher level than that of the normal (P < 0.005). The portal pressure of the spl-2W rats was similar to that of the PVL-2W and PVL-4W rats but higher than that of N-4W rats (15.2 + 4.2,14.0 + 2.7, and 15.2 f 2.7 vs 7.8 + 1.1 mm Hg; P < 0.005). The portal pressure of the spl-GM rats was significantly higher than that of the PVL-GM rats (17.1 * 6.4 vs 12.7 + 2.0 mm Hg; P < 0.05). Though SAL did not show any immediate decrease in portal pressure (14.0 f 2.7 mm Hg vs 14.6 -+ 1.4 mm Hg; P > 0.05), a significant decrease of portal pressure (15.2 f 2.7 mm Hg vs 12.0 + 1.9 mm Hg; P < 0.05) was noted in the SAL-2W rats. The decrease in portal pressure was more obvious (12.7 ? 2.0 vs 10.5 + 2.6 mm Hg; P < 0.05) in SAL-GM rats. The portal pressure in these rats was statistically lower than in PVL-GM rats (P < 0.05). Significant splenomegaly was noted in the PVL and SAL rats as compared with the normal controls, but the weight of the spleen in the SAL rats was lower than in the PVL rats (P -C 0.05). DISCUSSION
A sustained portal hypertension could be r ;producibly induced by PVL in rats [lo]. The portal pressure rose
LIN
Results
BW LW LW/BW (x10-2) SW SW/SW (x10-3)
of Splenectomy
SHAN:
PORTAL
and Splenic
HYPERTENSION
TABLE
1
Arterial
Ligation
623
IN RATS
on Portal
N-6M (n = 12)
PVL-PW (n = 12)
PVL-4w (n = 12)
PVL-GM (n = 12)
270 f 5 10.60 f 1.15
351 f 35 11.17 f 0.82
525 + 48 16.04 + 2.45
315 + 16 9.41 -t 1.61
356 + 39 9.64 + 1.96
542 -+ 24 13.59 t 0.64
352 + 29 9.55 * 1.2
449 k 44 10.06 t 1.95
324 + 43 12.21 + 1.59
493 k 57 11.26 + 1.80
3.84 0.97 + 0.09
3.18 0.95 + 0.07
3.06 0.95 i- 0.08
2.99 1.76 t 1.61
2.71 1.33 i 0.22
2.59 1.69 t 0.16
2.71 1.88 + 0.60
2.01 1.83 i 0.34
3.55 1.39 2 0.36
2.99 1.24 t 0.23
3.51
2.70
1.80
5.59 14.0 i 2.7 11.0 ?I 3.0* 14.6 k 1.4#
3.71
3.22
5.98
6.13
4.05
2.52
15.2 T 2.7
12.7 _t 2.0**
15.2 i 4.2@
17.1 + 6.4***
12.0 + 1.9##
10.5 f 2.6###
7.7 AI 0.8
Spl-GM (n = 12)
Rats
N-4W (n = 12)
7.8 + 1.1
Spl-PW (n = 12)
Hypertensive
N-l (n = 12)
6.0 f 0.5
PP
AND
SAL-SW (n = 12)
SAL-GM (n = 12)
Note. BW: body weight, g. LW: liver weight, g. SW: spleen weight, g. PP: portal pressure, mm Hg. N-l: normal rats of 2 months old. N-4W: N-l rats kept for 4 weeks more. PVL-PW: 2 weeks after portal vein ligation. Spl-2W: 2 weeks after splenectomy on the PVL-2W rats. SAL-PW: 2 weeks after splenic arterial ligation on the PVL-2W rats. * Spl-imm, P < 0.01 compared with PVL-2W. # SAL-imm, I’ > 0.1 compared with PVL-2W. ** P < 0.05 compared with N-6M; P < 0.01 compared with PVL-SW. @P > 0.1 compared with PVL-4W. *** P -C 0.05 compared with PVL-GM; P < 0.005 compared with N-6M. ## P < 0.005 compared with N-4W; P c 0.05 compared with PVL-4W. ### P < 0.05 compared with PVL-GM; P < 0.005 compared with Spl-GM, P < 0.05 compared with N-6M.
immediately after PVL and had been maintained thereafter, Sikuler and Groszmann [lo] stated that the mechanisms which contributed to the development and maintenance of portal hypertension change with time. The immediate rise in portal pressure after PVL was attributable to the abrupt increase in portal venous resistance. In addition to increased portal venous resistance, a hyperdynamic circulatory derangement with splanchnit hyperemia (or increased portal venous inflow) per-
20.00. 3 I
H
normal
-
PVL-only
w
PVL-spl
0-0
PVL-SAL
rat*
5 0 15.00
10.00
5.00
0.00
+ 0
2w
4W
6M TIME
FIG. 1. Effects of splenectomy and splenic artery ligation porta pressure in portal hypertensive rats.
on the
haps played an even more important role in the maintenance of short-term portal hypertension. This hyperdynamic circulation, however, was no longer present in later stages of portal hypertension. The portal pressure in this stage was primarily maintained by an increased resistance to the portal venous inflow. In our study, portal pressure of PVL-GM rats was significantly lower than that of PVL-2W rats, while it was still much higher than that of the normal controls. However, in their report [lo], Sikuler and Groszmann noted that the pressure in the long-term portal hypertensive rats was lower, but not statistically significant, than in the short-term group. There are still no solid experimental data to satisfactorily explain this phenomenon. However, it can be speculated that the force of hemodynamic homeostasis may have led the pressure to settle at a new, lower, and physiologically acceptable level in the long-term portal hypertensive rats. Regardless of what the pathophysiology of portal hypertension is [12-151, in this study we have observed that SAL was effective in reducing the portal pressure following PVL, while splenectomy was ineffective. This was most clearly demonstrated by the fact that splenectomy only led to a 20-set drop in the portal pressure, which bounced back and became even higher than the presplenectomy level. The drop obviously was due to a very transient cut-off in splenic flow, which was more than adequately compensated by an increase in the mesenteric blood flow [ 161. The spleen, moreover, may act as an effective circulatory reservoir with its potential to develop extensive collateral pathways [ 171. In addition, it is possible that the spleen may be capable of producing some still unidentified humoral factors which could regu-
624
JOURNAL
OF SURGICAL
RESEARCH:
late the portal pressure. Removal of the spleen will effectively eliminate these advantages from the body and leave the portal pressure unharnessed. It is interesting to note that our experimental data could be invoked to explain the observations made by Koyama et al. in a clinical setting [16]. They observed that the portal pressure in patients, who had undergone splenectomy for idiopathic portal hypertension several years before, was distinctly increased. They speculated that this was due to increased superior mesenteric blood flow and/or increased block in the liver in these patients. When SAL was performed in the PVL-2W rat, no significant decrease in portal pressure was observed immediately but was 2 weeks after the procedure. A further drop in the portal pressure was observed in the ensuing 6 months. Three explanations could be offered for this observation. One is that SAL cut off the splenic arterial inflow, which has known to be increased in portal hypertension [7, 81. In their report, Witte et al. noted that ligation of the splenic artery close to the splenic hilum would interrupt the splenic inflow and promoted splenic infarction and thereby reduction in functional splenic mass [ 17,181. This phenomenon was also clearly shown in our study; the splenic weight of the splenic artery ligated group was statistically lower than that of the unligated group. Second, with the ligated spleen in situ, the porto-systemic shunts in and around the spleen were left intact, which could lower the portal pressure. Moreover, these porto-systemic venous channels no longer draining a hyperdynamic splenic circuit, which was interrupted by splenic artery ligation, may be converted into an escape route for mesenteric venous blood entering the portal system under higher pressure [17, 181. Third, since in SAL rats, the spleen was not removed, its above-mentioned regulatory functions on portal pressure were still at work. Therefore, the pressure determined in these animals was significantly lower than that of splenectomized rats. The spleen, therefore, appears to have a modulating function on the portal pressure in this animal model, and ligation of splenic artery may bring about a better effect for portal hypertensive rats than splenectomy. We concluded that in the long run splenectomy would not lower the portal pressure of portal hypertensive rats, while ligation of the splenic artery would.
ACKNOWLEDGMENTS The authors thank H. Y. Chang for statistical support, H. H. Hou for typing the manuscript, and Prof K. Y. Huang for reviewing and revising the manuscript.
VOL.
53, NO. 6, DECEMBER
1992
REFERENCES 1.
Dumont, A. E., Amorosi, E., and Stahl, W. M. Significance of splenomegaly in patients with hepatic cirrhosis and bleeding esophageal varices. Ann. Surg. 171: 522, 1971. 2. Dumont, A. E., Berman, I. R., Stahl, W. M., and Ring, S. M. Significance of an enlarged splenic artery in patients with bleeding varices. Ann. Surg. 175: 466, 1972. 3. Witte, C. L., Witte, M. H., Bair, G., Mobley, W. P., and Morton, D. Experimental study of hyperdynamic vs. stagnant mesenteric blood flow in portal hypertension. Ann. Surg. 179: 304, 1974. 4. Hassab, M. A. Gastroesophageal decongestion and splenectomy in the treatment of esophageal varices in bilharzial cirrhosis: Further studies with a report on 355 operations. Surgery 61: 169, 1967. 5. Sugiura, M., and Futagawa, S. A new technique for treating esophageal varices. J. Thorac. Cardiouasc. Surg. 66: 677, 1973. 6. Sugiura, M., and Futagawa, S. Esophageal transection with paraesophagogastric devascularizations (the Sugiura Procedure) in the treatment of esophageal varices. World J. Surg. 8: 673,1984. 7. Vorobioff, J., Bredfeldt, J. E., and Groszmann, R. J. Hyperdynamic circulation in portal-hypertensive rat model: A primary factor for maintenance of chronic portal hypertension. Am. J. Physiol. 144: G52, 1983. 8. Groszmann, R. J., Vorobioff, J., and Riley, E. Splanchnic hemodynamics in portal hypertensive rats: Measurement with r-labeled microspheres. Am. J. Physiol. 242: G156,1983. 9. Sikuler, E., Kravetz, D., and Groszmann, R. J. Evolution of portal hypertension and mechanisms involved in its maintenance in a rat model. Am. J. Physiol. 248: G618,1985. 10. Sikuler, E., and Groszmann, R. J. Hemodynamic studies in longand short-term portal hypertensive rats: The relation to systemic glucagon level. Hepatology 6: 414, 1986. 11. Lin, P. W., and Shan, Y. S. Effects of splenectomy on portal pressure in portal hypertensive rats. J. Surg. Assoc. ROC 23: 208, 1990. 12. Benoit, J. N., and Granger, D. N. Splanchnic hemodynamics in chronic portal hypertension. Semin. Liver Dis. 6: 287, 1987. 13. Benoit, J. N., Womack, W. A., Hernandez, L., and Granger, D. N. “Forward” and “backward” flow mechanisms of portal hypertension, relative contributions in the rat model of portal vein stenosis. Gastroenterol. 89: 1092, 1985. J. A., Harper, S. L., Kvietys, P. R., 14. Benoit, J. N., Barrowman, and Granger, D. N. Role of humoral factors in the intestinal hyperemia associated with chronic portal hypertension. Am. J. Physiol. 247: G486, 1984. of flow and resis15. Sikuler, E., and Groszmann, R. J. Interaction tance in maintenance of portal hypertension in a rat model. Am. J. Physiol. 250: G205, 1986. 16. Koyama, K., Ouchi, K., and Sato, T. Development mechanism of portal hypertension in idiopathic portal hypertension. In K. Okuda, and M. Omata (Eds.), Idiopathic Portal Hypertension. Univ. of Tokyo Press, 1983. Pp. 231-240. 17. Witte, C. L., Witte, M. H., Renert, W., O’Mara, R. E., and Lilien, D. L. Splenic artery ligation in selected patients with hepatic cirrhosis and in Sprague-Dawley rats. Surg. Gynecol. Obstet. 142: 1,1976. 18. Witte, C. L., Corrigan, J. J., Witte, M. H., and O’Mara, R. E. Splenic artery ligation in experimental hypersplenism. Surg. 80: 581,1976.
