Ultrasonographic Diagnosis of Portal Vein Cavernous Transformation in Children

Transcription

1 Ultrasonographic Diagnosis of Portal Vein Cavernous Transformation in Children Bernardo Frider, MD*, Ana Maria Marin, MOt, Alberto Goldberg, MD:t: The aim of this work was to evaluate the efficiency, sen sitivity, and specificity of ultrasonographic findings in the diagnosis of portal vein cavernous transformation and secondary splanchnic hypertension in children. Twenty patients with esophageal varices on endoscopic sclerotherapy, with ages ranging from 1 to 15 years (mean of 7.5), were compared with age ~ matched normal asymptomatic controls. In 17 cases, diagnosis was confirmed by splenoportography. Ultrasonographic portal vein features, respiratory variations in splenic and superior mesenteric veins, lesser omentum to aorta diameter ratio and spleen size, as well as the presence of vessels in the lesser omentum and of spontaneous anas tomoses, were analyzed in both groups. The efficiency of ultrasonography to establish a diagnosis of portal vein cavernous transformation reached %. Splenic and superior mesenteric vein respiratory variation was negligible in patients and inspiration versus expiration U diameter differences ranged from 1 to 4 mm (mean of 24) in controls (p <.001). Mean lesser omentum to aorta ratio was 7 in patients versus.9 in controls (p <.001). Splenomegaly was found in all but two cases; lesser omentum vessels were visualized in nine cases, whereas five presented spontaneous splenorenal anastomoses and gallbladder varices were observed in five patients. Thus, ultrasonography allowed portal vein cavernous transformation to be diagnosed throughout; portal hypertension was also detected. It may be concluded that ultrasonography, either alone or associated with endos copy, provides a reliable method for portal vein cavernous transformation diagnosis. Angiography should be reserved for preoperative or controlled hemodynamic studies. KEY WORDS: portal hypertension, portal vein cavernous transformation, ultrasonography. (/ Ultrasound Med 8:445, 1989) ltrasonography (US) is a well-proven diagnos ~ tic method to evaluate changes in splanchnic circulation induced by portal hypertension (PH). Thus, dilated splanchnic veins, 1 collaterals,l recanalized umbilical vein, 3 respiratory variations of the splenoportal axis, 4 6 and lesser omentum enlargement7 may be readily visualized. Also, surgical anastomosis permeability 8 may be controlled, portal flow studied by the Doppler effect, and drug therapeutic action evaluated.9 Given its innocuousness. easy repetition, and wealth of information provided, US is ideal for diagnosis and follow-up of PH in children,'' most commonly due to portal vein cavernous transformation (PCT). 10 The aim of the present work was to evaluate the application of US in the diagnosis of PCT and subsequent hypertension in a pediatric population. Received July, 1988, from the Department of Internal Medicine, Argerich Hospital, Buenos Aires, Argentina; and the Departments of t5urgery and tradiology. Ricardo Gutierrez Children's Hospital, Buenos Aires, Argentina. Revised manuscript accepted for publication January 12, Address correspondence and reprint requests to Dr. Frider: Sal guero 2601, (1425) Buenos Aires, Argentina. SUBJECTS AND METHODS 1989 by the American Institute of Ultrasound in Medicine Two pediatric groups were studied: group I consisted of normal asymptomatic children whose ages ranged from 2 to 13 years (mean± SO: 7.2 ± 3.35 years) and J Ultrasound Med 8: , /89/$3.50

2 446 J Ultrasound Med 8: , 1989 PORTAL VEIN CAVERNOUS TRANSFORMATION Table 1: Sex and Age Distribution in Patients with Portal Vein Cavernous Transformation (PCT) and in Controls Group Number Male Female Age range (yr) Mean ± SO (yr) Patients Controls Total ± ± 3.35 group II consisted of children with PCT aged from 1 to 15 years (mean ± SO: 7.5 ± 3.50 years). Seventeen group ll patients had been diagnosed by splenoportography and one by surgery, whereas the re maining two were included after US study. All sought advice for digestive hemorrhage due to esophageal varices and were under treatment with endoscopic sclerotherapy in the surgery service at Ricardo Gutierrez ChiJ, dren's Hospital, Buenos Aires (Tables 1 and 2). In both groups, blind study US was performed by means of a high-resolution, real time Atoka 248 (Coro metrics; Wallingford, CT) device with 3.5-MHz and 5-MHz sector transducers, whereas some patients were reexamined by a real-time Aloka 280 model with a 3.5MHz convex transducer. Studies were carried out after at least 3 hours of fas ting. All cases were examined in supine position by the anterior pathway and by intercostal pathway when meteorism prevented adequate portal vein visualization. At US examination, observation was focused on visualization and features of the main portal, superior mesenteric, and splenic veins; and respiratory variation of the splenic and superior mesenteric veins. Splenic vein diameter at inspiration and expiration was determined at the intersection of the superior mesenteric artery in a transverse section. The diameter of the superior mesenteric vein was measured at inspiration and expiration in a sagittal section roughly 1 em from the source of the portal vein. Lesser omentum and aorta diameters were determined in a sagittal section I em from the outlet of the celiac axis and the lesser omentum to aorta diameter ratio was calculated. Also, the presence of vessels in the lesser omentum was investigated; spleen dimensions were evaluated; spontaneous splenorenal anastomoses were searched for; and remaining abdominal organs were studied to rule out other congenital anomalies associated with PCT.10 The indirect nosologic method 11 was employed for diagnosis. Sensitivity, specificity, and efficiency of each finding at US were analyzed with 95o/o confidence intervals Table 2: Splenoportography in Cases of PCT Previous Spleen agenesis Not performed Figure 1 Vessels in lhi.ckened lesser omentum (23 em). Aorta (sl1ort arrow) measures 1 em; lesser omentum to aorta ratio is 23 (lo11g arrow indicates esophagogastric junction). RESULTS In all group I normal children, features of the main portal vein and its intrahepatic branches were visual ized and analyzed. Likewise, the main portal vein and its branches were evaluated in all group II patients. Both Figure 2 Arrow indicates gallbladder varices.

3 J Ultrasound Med 8:445 ~ 449, FRIDER ET AL The portal vein proved pathologic in all group II patients, with trunk replacement by a network of vessels in 17 cases and by slender tortuous vessels within an echogenic pattern in the remaining 3 cases (Fig. 4, A and B). DISCUSSION Figure 3 Spontaneous splenorenal anastomosis (heavy arrow indicates left renal vein; slender arrows show vessels between spleen and kidney). splenic and superior mesenteric veins were observed throughout in group I and the superior mesenteric vein in only 11 group II patients. Respiratory variation (inspiration versus expiration size difference) of the splenic and superior mesenteric veins ranged from 1 to 4 mm (mean ± SO: 2.4 ±.8) in group I and was negligible in group II (p <.001). The lesser omentum was visualized and its relation to the aorta determined in all group I children, but in group II it was only observed in 13 cases. Average lesser omentum to aorta ratio was.9 ±.25 with a range of.54 to 1.18 in group l compared with 2.07 ±.51 with a range of 1.18 to 3.40 in group II (p <.001). Neither vessels in the lesser omentum nor spontaneous anasto moses were observed in group I children, whereas in group II, lesser omentum vessels were visualized in nine cases, gallbladder varices in five cases, and spontaneous splenorenal anastomoses in five cases (Figs. 1, 2, and 3) (Table 3). Table 3: Visualization of Vessels and Lesser Omentum in Patients with PCf Visualization Number Percentage (%) Portal vein malformation Splenic vein Lesser omentum Superior mesenteric vein Lesser omentum vessels Gallbladder varices Spontaneous anastomoses Pediatric PH presents some special features owing to the great variety of etiologies, apparent surgical difficulties due to vessel size, hypertension tolerance lasting several years, and the possibility of spontaneous regression with growth. 7 Despite the fact that PCf is the most common cause of hypertension in childhood, its incidence was barely 1: 50,000 in first time consultations at the Ricardo Gu tierrez Children's Hospital from April 1983 to July Real-time US allows easy visualization of changes to follow the vessel pathway, the main difficulty being the presence of meteorism, which hinders the detection of all relevant signs (as shown in Table 3). Such meteorism is not uncommon in hypertensive patients and explains the occasional need for repeated US examination. Portal vein features recorded in our work agreed with descriptions by other authors The portal bifurcation may be replaced by an echogenic structure with multiple small tortuous vessels (Fig. 5). The portal trunk may appear as a network of vessels (Fig. 4A) or as slender tortuous vessels within an echogenic structure (Fig. 48). Lesser omentum thickening, due to lymphatic and venous edema, may increase its diameter to three or four times that of the aorta In our series, the efficiency of ultrasound was 97%, with 92.3% sensitivity and % specificity (see Table 4). The visualization of lesser omentum vessels supplements PH diagnosis and enables the presence of esophageal varices to be predicted with 93% reliability 15 (Fig. 1). The lack of respiratory variation in the superior mesenteric and splenic veins confirms previous studies on adults with cirrhosis, which show that such absence correlates with the presence of PH and esophageal varices. In our work, effi ciency, sensitivity, and specificity all reached %. AI though the prevalence of ectopic varices in PH patients is not well-established, it may be estimated as 1% to 3% in cirrhosis cases and % to 30% in those with extrahepatic PH. 19 In our series, gallbladder varices were found in 25% (5 cases), confirming that this location is commonly associated to extrahepatic PH, so that careful ob ~ servation of the gallbladder is advisable in case of clinical suspicion. Extraesophagogastric portosystemic anastomoses have been reported in the literature with varying prevalence according to the diagnostic methods

4 448 J Ultrasound Med 8: , 1989 PORTAL VEIN CAVERNOUS TRANSFORMATION A B Figure 4 A, Portal vein cavernous transformation. Replacement of single portal trunk by a network of tortuous vessels. B, Vessels replacing the portal trunk are seen within an echogenic pattern. FigureS Pathologic portal vein bifurcation (arrows indicate a network of vessels at the bifurcation site). employed.1.21m 6 Transhepatic portography seems to provide the greatest sensitivity in detecting such anastomoses. In our population, all five splenorenal anasto moses were seen both by US and by splenoportography, whereas two further thoracohepatic anastomoses were visualized by angiographic studies, but not by US. The actual capability of such anastomoses to decompress the hypertensive portal system is still a matter of controversy. 26 In cases without esophageal varices, which ef fectively serve as shunting mechanisms, collateral circulation proceeded by way of a single vessel of great size Not all of our patients presenting spontaneous anastomoses had esophageal varices at the time of US examination: out of the seven cases with shunting, five had splenorenal and two had thoracohepatic anastomoses. All had bleeding, grade IV esophageal varices on starting endoscopic sclerotherapy, which reduced in sizes to grades I and II, whereas one case had no esophagogastric varices at the time of US examination (two years on sclerotherapy) and was the only group II case in Table 4: Efficiency, Sensitivity, and Specificity of Ultrasonographic Findings in Children with PCT U1trasonographic signs Pathologic portal vein Respiratory variation in superior mesenteric and splenic veins Splenomegaly Lesser omentum to aorta ratio Lesser omentum vessels Efficiency (o/o) Sensitivity (%) Specificity (%)

5 JUltrasound Med 8: , 1989 which the lesser omentum to aorta ratio proved normal (1.18). Despite the limited number of cases studied, the presence of spontaneous anastomoses seems to have prognostic value as regards the outcome of sclerotherapy, a hypothesis that may be evaluated by longitudinal follow-up of these patients. In this connection, US affords a reliable tool for PCT follow-up in children, given the frequency with which hemorrhage due to varices remits with bodily growth, presumably attributable to the development of anastomoses by recanalization of closed or partially occluded embryonal canals/ so that sclerosis or embolization of esophageal varices would foster this process. 25 The presence of splenomegaly in % of cases supports US examination as first study in children to rule out PH as an etiologic factor. Together with endoscopy, US is suitable to diagnose both PH and PCT in almost all patients, whereas angiography should be reserved for preoperative or controlled hemodynamic studies. REFERENCES Kane RA, Katz SG: The spectrum of sonographic findings in portal hypertension. A subject review and new observations. Radiology 142:453, 1982 Schabel S, Rittenberg G, Cunningham J, et al: "Bulls eyes" falciform ligament: A sonographic finding of portal hypertension. Radiology 136:157, 1980 Fakhry J, Gosink BB, Leopold GR: Recanalized umbilical vein due to portal vein occlusion: Documentation by sonography. AJR 137:410, 1981 Bolondi L, Gandolfi L, Arienti V, et al: Ultrasonography in the diagnosis of portal hypertension. Diminished response of portal vessels to respiration. Radiology 142:167, 1982 Kurol M, Forsberg L: Ultrasonographic investigation of respiratory influence on diameters of portal vessels in normal subjects. Acta Radio) [DiagnJ (Stockh) 27:675, 1986 Rabinovicci N, Navot N: The relationship between respiration, pressure and flow distribution in the vena cava and portal and hepatic veins. Surg Gynecol Obstet 151:753, 1980 Bernard 0, Alvarez F, Brunelle F, et al: Portal hypertension in children. Clin Gastroenterol 14:33, 1985 Zoli M, Marchesini G, Cordiani MR, et al: Echo-Doppler measurement of splanchnic blood flow in control and cirrhotic subjects. JCU 14:429, 1986 Zoli M, Marchesini G, Marzochi A, et at: Portal pressure changes induced by medical treatment: US detection. Radiology 155:763, 1985 FRIDER ET AL Alvarez F, Bernard 0, Brunelle F, et al: Portal obstruction in children. I. Clinical investigation and hemorrhage risk. J Pediatr 103:696, Wulff H: Rational Diagnosis and Treatment. Oxford, Blackwell, Galen RS: Predictive value and efficiency of laboratory testing. Pediatr Clin North Am 27:861, Tablas Cientificas Geigy, ed 6. Baste (Switzerland), JR Geigy Co, 1965, pp Patient admission data from the Department of Statistics, Ricardo Gutierrez Children's Hospital, Buenos Aires, Brunelle F: Portal hypertension. In: Kalifa G (ed): Pediatric Ultrasonography, New York, Springer Verlag, in press 16. Alagitle D: Hypertension portale de I'enfant. Arch Fr Pediatr 43:441, Zoli M, Dondi C, Marchesini G, et al: Splanchnic vein measurement in patients with liver cirrhosis: A case control study. J Ultrasound Med 4:641, Cottone M, Damico G, Maringhini A, et at: Predictive value of ultrasonography in the screening of non ascitic cirrhotic patients with large varices. J Ultrasound Med 5:189, Lebrec D, Benhamou JP: Ectopic varices in portal hypertension. Clin Gastroenterol 14:105, Di Candio G, Campatelli A, Mosca F, et al: Ultrasound detection of unusual spontaneous portosystemic shunts associated with uncomplicated portal hypertension. J Ul trasound Med 4:297, Rousselot M, Moreno AH, Pankew F: Studies in portal hypertension. IV. The clinical and physiopathologic sigr nificance on self established (non surgical) portal sys ~ temic venous shunts. Ann Surg 150:384, Warren WD, Zeppa R, Fomon Jj: Selective trans-splenic decompression of gastroesophageal varices by distal splenorenal shunt. Ann Surg 166:437, juttner HU, Jenney JM, Ralls PW, et al: Ultrasound dem onstration of portosystemic collaterals in cirrhosis and portal hypertension. Radiology 142:459, Dach JL, Hill MC, Pelaez JC, et al: Sonography of hypertensive portal venous system. Correlation with arterial portography: AJR 137:511, Subramanyan BR, Balthazar Ej, Madamba MR, et al: Sonography of portal systemic venous collaterals in portal hypertension, Radiology 146:161, Nunez D Jr, Russell E, Yrrizarry J, et al: Portosystemic communications studied by transhepatic portography. Radiology 127:75, Wexler MJ, Mac Lean LD: Massive spontaneous portal systemic shunting without varices. Arch Surg 110:995, 1975