0Journal of Medical Ultrasound (2014) 22, 66-70
Available online at www.sciencedirect.com
ScienceDirect
journal homepage: www.jmu-online.com
REVIEW ARTICLE
Hepatic Portal Venous Gas: Review of the Literature and Sonographic Implications
Chun-Yen Huang 1, Jen-Tang Sun 1, Kuang-Chau Tsai 1, Hsiu-Po Wang 2, Wan-Ching Lien 3*
CrossMark
1 Department of Emergency Medicine, Far Eastern Memorial Hospital, New Taipei City, Taiwan,
2 Department of Internal Medicine, National Taiwan University and National Taiwan University Hospital, and3 Department of Emergency Medicine, National Taiwan University and National Taiwan University Hospital, Taipei, Taiwan
Received 18 June 2013; accepted 15 April 2014 Available online 17 June 2014
KEY WORDS
hepatic portal venous gas,
ischemic bowel, pneumatosis
intestinalis, ultrasound
To review the clinical data and sonographic implications of hepatic portal venous gas (HPVG), we collected 447 cases of HPVG from the MEDLINE database and the National Taiwan University Hospital, Taipei, Taiwan database and analyzed etiology, radiographic findings, diagnostic tools, and outcome of all the cases. Among them, 61% of cases were nonischemic bowel disease. Mortality rates are significantly influenced by bowel ischemia and pneumatosis intestina-lis. Ultrasound-diagnosed HPVG was more frequently reported after 1980 and associated with etiology. In conclusion, the number of cases with ultrasound-diagnosed HPVG is increasing. The etiology of HPVG should be searched and coincident imaging findings, including pneumatosis intestinalis, should be evaluated.
© 2014, Elsevier Taiwan LLC and the Chinese Taipei Society of Ultrasound in Medicine. All rights reserved.
Introduction
Hepatic portal venous gas (HPVG) is considered an ominous sign, referring to an abdominal catastrophe. It was first
Conflicts of interest: The authors did not receive any financial support and have no conflicts of interest.
* Correspondence to: Dr Wan-Ching Lien, Department of Emergency Medicine, National Taiwan University Hospital, Number 7, Chung-Shan South Road, Taipei 100, Taiwan.
E-mail address: temer17@yahoo.com.tw (W.-C. Lien).
described in 1955 by Wolfe and Evans [1] in six infants dying from necrotizing enterocolitis. Four of them had a vascular pattern of gas in the liver seen on plain films prior to death. In 1978, Liebman et al [2] reported 64 cases of HPVG with a mortality rate of 75%. Moreover, the mortality rate might be as high as 90% in patients with mesenteric artery occlusion when complicated with HPVG [3]. Recently, HPVG is also found in a more extended spectrum of clinical settings as a result of improvements in imaging modalities such as diverticulitis, ventral hernia without ischemic bowel [4], hypertrophic pyloric stenosis [5], chronic obstructive pulmonary disease (COPD) [6], iatrogenic gastric dilatation [7],
http://dx.doi.Org/10.1016/j.jmu.2014.04.008
0929-6441/© 2014, Elsevier Taiwan LLC and the Chinese Taipei Society of Ultrasound in Medicine. All rights reserved.
ileus [8], blunt abdominal trauma without ischemic bowel [9], extracorporeal shockwave lithotripsy [10], Crohn disease [11], appendicitis [12], liver abscess [13], superior mesenteric artery syndrome [14], enterovascular fistula [15], perforated gastric ulcer [16], jejunal feeding tube [17], nonocclusive mesenteric ischemia [18], and hydrogen peroxide ingestion [19]. Not all the conditions require surgical intervention. Benign HPVG is that which is caused by a condition other than ischemic bowel disease. Additionally, the mortality rate of HPVG decreases to 25% [20].
Sonography is a noninvasive, easy-to-perform, and readily accessible diagnostic tool. HPVG detected by ultrasound (US) was first reported by Fataar et al [21] in 1986. They described a 55-year-old man with sigmoid diverticu-litis whose ultrasonogram showed some small hyperechoic foci in the region of the dome of the liver and then throughout the liver 2 days later, referring to HPVG. US is frequently used for diagnosing HPVG and has a well-established role in detecting the presence of HPVG and distinguishing it from gas in the bile ducts [22—24]. Furthermore, HPVG detected by US can predict poor outcome in out-of-hospital patients in cardiac arrest [25]. However, previous studies were limited by case reports and sonographic implications were not thoroughly investigated.
This review analyzes the clinical data from cases of this entity from the literature, including 19 patients we treated, and investigates sonographic implications of HPVG.
Patients and methods
We performed a search of the MEDLINE database (National Library of Medicine, Bethesda, MD, USA) for HPVG, using "portal venous gas" and "gas in the portal vein" as keywords. Other articles were identified by cross referencing. All literature is in English. The search spanned from 1955 through September 2012. Four hundred and twenty-eight cases identified from the literature search were included in the analysis. Cases with HPVG at postmortem or cardiac arrest were excluded. We reviewed etiology, radiographic findings, diagnostic tools, and outcome of all the cases. These cases, along with the 19 cases in which the diagnosis was made at the National Taiwan University Hospital, Taipei, Taiwan were put into analysis. Statistical analyses were performed using SAS version 9.2 software. Student t test was performed for continuous data, as well as the Chi-square test for categorical data. A p value < 0.05 was considered significant.
Results
During this review period, there were 447 patients in whom HPVG was diagnosed, including 19 cases that were treated. Among them, 257 patients (61%) were male. Their clinical data are summarized in Table 1. The age ranged from 0 years to 100 years and the mean was 54.6 ± 24.7 years. HPVG resulting from ischemic bowel was frequent in patients younger than 1 year or older than 65 years. However, the age of 14 patients and the sex of 28 patients cannot be identified in the literature.
The mortality rate caused by ischemic bowel is higher than the rate caused by nonischemic bowel, especially in
Table 1 Characteristics of the 447 patients with hepatic portal venous gas.
Ischemic etiology Nonischemic etiology Total P
Total number 176 (39) 271 (61) 447
Age 1 —65 y 64 (36) 161 (59) 225 (50)
Age <1 or >65 y 103 (59) 105 (39) 208 (47) <0.0001
Unknown age 9 (5) 5(2) 14(3)
Mortality 110 (63) 48 (18) 158 (35) <0.0001
Mortality in age 37 (34) 27 (56) 64 (41)
1 —65 y
Mortality in age 65 (59) 20 (42) 85 (54) 0.0153
<1 or >65 y
Mortality in 8 (7) 1 (2) 9(5)
unknown age
Data are presented as n (%).
patients younger than 1 year or older than 65 years. Among nonischemic bowel etiologies, the most common cause of HPVG is iatrogenic-related (19%, N = 54), such as enema, endoscopy, and interventional procedures, followed by diverticulitis (10%, N = 29) and ileus (9%, N = 26). Other etiologies include hollow organ perforation (N = 17), inflammatory bowel disease (N = 13), blunt abdominal trauma without ischemic bowel (N = 11), idiopathic cause (N = 11), acute pancreatitis (N = 9), shock (N = 9), liver transplantation (N = 7), enterocolitis (N = 7), chemotherapy (N = 6), cholangitis (N = 4), gastrointestinal volvulus (N = 4), gastrointestinal hernia (N = 4), superior mesentery artery syndrome (N = 4), chronic obstructive pulmonary disease (N = 4), gastric ulcer (N = 4), emphysematous gastritis (N = 3), hypertrophic pyloric stenosis (N = 3), liver abscess (N = 2), cholecystitis (N = 2), intraabdominal abscess (N = 2), appendicitis (N = 2), acute fulminant hepatitis (N = 2), emphysematous pyelonephritis (N = 2), graft versus host disease (N = 2), hydrogen peroxide ingestion (N = 2), and infrequent causes (total number 26, N = 1 respectively).
With the advancement of imaging modalities, the diagnostic tools for HPVG have changed. Prior to 1990, there were only four reported cases of HPVG diagnosed by US, and HPVG was frequently detected by plain films. After 1990, there were 80 cases diagnosed by US (Table 2). Although a certain percentage of the cases of HPVG needed CT to confirm the diagnosis, some cases of HPVG were found only by US.
Table 3 shows the relationship between mortality rate and HPVG with/without pneumatosis intestinalis (PI). The presence of PI in patients of HPVG was associated with ischemic bowel and poor outcome.
Additionally, our review focused on sonographic implications of HPVG (Table 4). The etiologies of US-diagnosed HPVG were frequently associated with benign HPVG.
Discussion
HPVG was thought to be a fatal sign and predictive of bowel ischemia [8]. However, we can find that HPVG caused by
C.-Y. Huang et al.
Table 2 Changing trends in diagnostic tools for hepatic portal venous gas.
Diagnostic tool Prior to 1990 After 1990
(n = 87) (n = 360)
CT alone 6 254
Plain film alone 75 11
US alone 3 38
Plain film + CT 1 13
Plain film + US 1 2
Plain film + US + CT 0 2
US + CT 0 38
ERCP 0 1
Surgery 1 0
Barium enema 0 1
CT = computed tomography; ERCP = endoscopic retrograde cholangiopancreatography; US = ultrasound.
nonischemic bowel disease was increasingly reported over the past 20 years. Analytical results show the pooled mortality rate is as low as 35%, which can be explained to be buffered by the lower mortality rate of HPVG caused by nonischemic bowel disease. Additionally, among non-ischemic bowel-related HPVG, an iatrogenic cause is the most common.
Two mechanisms were proposed for the development of HPVG: (1) microbe-derived gas production, such as appendicitis or diverticulitis, and (2) absorbed intraluminal air caused by an impaired epithelial barrier or increased intraluminal pressure, such as caustic ingestion, superior mesenteric artery syndrome, or ileus [26]. In our review, we also found some cases of HPVG that are difficult to explain by these two mechanisms, such as HPVG caused by COPD or emphysematous pyelonephritis. In the patient with COPD, shortness of breath and dyspnea may cause much gas to be ingested into the gastrointestinal tract and bowel disten-tion, and mucosal tears allowing the entry of air into the intramural part of bowel that then migrates into the portal circulation, which contributes to the development of HPVG
[27]. In a patient with emphysematous pyelonephritis, air in the portal vein may be caused by aerogenic bacteremia
[28], portosystemic shunt [29], or an inflamed mesenteric venous bed from colitis caused by inflammatory tracking from pyelonephritis [30]. However, a certain percentage of HPVG was idiopathic [4].
Table 3 Relationship between mortality and hepatic portal venous gas with/without pneumatosis intestinalis.
HPVG HPVG p
with PI without PI
(n = 131) (n = 316)
Etiology
Ischemic bowel 84 (64.1) 92 (29.1)
Nonischemic bowel 47 (35.9) 224 (70.9) <0.0001
Mortality 64 (48.9) 94 (29.7) 0.0001
Data are presented as n (%).
HPVG = hepatic portal venous gas; PI = pneumatosis
intestinalis.
Table 4 Etiology and mortality of 84 cases with hepatic portal venous gas diagnosed ultrasonographically.
Number
Etiology
Ischemic bowel 31
Nonischemic bowel 53
Mortality 24
Prior to 1990, plain film was frequently used to diagnose HPVG and is still the current standard imaging modality of choice in most emergency departments. On a supine plain abdominal radiograph, HPVG appears as branching, linear, radiolucent vessels that may extend from the region of the main portal vein toward the periphery of both hepatic lobes, especially in the edge of the liver within 2 cm of the liver capsule. However, plain film is not sensitive enough for HPVG [31]. Because of advancements in diagnostic tools, incidental findings of HPVG has occurred more often. In 1986, US was first used for diagnosing HPVG [21]. On realtime ultrasonography, HPVG is viewed as small, mobile, echogenic foci moving quickly in the direction of blood flow inside the lumen of the portal vein [32]. Doppler studies show these foci to produce audible crackles and sharp bidirectional spikes superimposed on the usual Doppler tracing of the portal vein. The main limitation of US is operator dependence [32]. On CT scan, HPVG appears as peripheral, subcapsular, branching hypodensities in the liver, predominantly in the left lobe because of its ventral location [33]. Otherwise, HPVG can be differentiated from pneumatobilia on CT scan because it tends to accumulate peripherally as a result of the centrifugal flow of portal venous blood as opposed to the centripetal flow of bile [34]. In one analysis of T-tube cholangiograms, Sisk found no patient in whom contrast material came to within 2 cm of the liver capsule, whereas normal portal venograms demonstrated the venous radicles extending almost to the periphery of the liver. Because of the nature of the flowing blood, HPVG is usually transient unless continued large amounts of gas are produced [34]. Although both CT and US are more sensitive than plain radiography for detection of HPVG, sometimes HPVG may only be seen on US [32,35,36]. This may be because HPVG is a transient phenomenon as a result of high blood solubility gas-like carbon dioxide or even higher sensitivity of ultrasound for a small amount of gas [36].
In the past time, the presence of HPVG and PI is predictive of bowel infarction. Mortality rates have been reported to exceed 75% in patients in whom PI and HPVG are both seen on plain films [2,37]. In the study by Wiesneret al [38], approximately 91% of patients with both HPVG and PI show transmural bowel infarction (N = 16). Wayne et al [4] studied 88 cases of HPVG or PI, and found that small bowel PI is an indicator of bowel ischemia (p = 0.04, odds ratio = 5.1, confidence interval = 1.2—21.4). In the study by Kernagis et al [39], nine of the 15 patients with PI had bowel ischemia and four of them had HPVG. The remaining six patients with PI had no bowel ischemia and none of them had HPVG. Khorram-Manesh and Oden [40], in a case
series, showed HPVG combined with PI does not influence mortality significantly (105 patients, p = 0.4, odds ratio = 1.48, confidence interval = 0.59—3.75). However, in our review, we find that when accompanied by HPVG, PI is significantly associated with bowel ischemia and mortality. The conflicting results may be because of different case numbers. To our knowledge, our review is the largest regarding HPVG.
Despite the contributions, our review has several limitations. First, publication bias exists in the included studies. HPVG is considered an ominous sign that journals may have a tendency to accept papers reporting a favorable outcome of HPVG, such as "benign HPVG". Therefore, the pooled mortality rate may not reflect real conditions. An actual percentage of ischemic bowel-related HPVG may not be reported in the literature. Second, the cases with HPVG diagnosed by US were still limited. However, US is operator dependent and sonographers may not be familiar with evaluation for ischemic bowel. In conclusion, the number of cases with US-diagnosed HPVG is increasing. However, the etiology resulting in HPVG should be carefully examined, especially ischemic bowel. Coincident presence of PI is associated with ischemic bowel and high mortality.
References
[1] Wolfe JN, Evans WA. Gas in the portal veins of the liver in infants; a roentgenographic demonstration with postmortem anatomical correlation. Am J Roentgenol Radium Ther Nucl Med 1955;74:486—8.
[2] Liebman PR, Patten MT, Manny J, et al. Hepatic—portal venous gas in adults: etiology, pathophysiology and clinical significance. Ann Surg 1978;187:281—7.
[3] Bloom RA, Lebensart PD, Levy P, et al. Survival after ultrasonographic demonstration of portal venous gas due to mesenteric artery occlusion. Postgrad Med J 1990;66: 137—9.
[4] Wayne E, Ough M, Wu A, et al. Management algorithm for pneumatosis intestinalis and portal venous gas: treatment and outcome of 88 consecutive cases. J Gastrointest Surg 2010;14: 437—48.
[5] Bhargava P, Parisi M. Gastric pneumatosis and portal venous gas: benign findings in hypertrophic pyloric stenosis. Pediatr Radiol 2009;39:413.
[6] Gorospe EC. Benign hepatic portal venous gas in a critically ill patient. Sci World J 2008;8:951—2.
[7] Bani-Hani KE, Heis HA. Iatrogenic gastric dilatation: a rare and transient cause of hepatic-portal venous gas. Yonsei Med J 2008;49:669—71.
[8] Kaufman D, Kang KH, Schwartzman A. Image of the month. Benign portal venous gas. Arch Surg 2008;143:803—4.
[9] Vercruysse GA, Adams SD, Feliciano DV. Computed tomo-graphic evidence of hepatic portal venous gas after blunt abdominal trauma does not necessitate surgery. Am Surg 2008;74:335—7.
[10] Rana AA, Sylla P, Woodland DC, et al. A case of portal venous gas after extracorporeal shockwave lithotripsy and obstructive pyelonephritis. Urology 2008;71:546. e5—7.
[11] Salyers Jr WJ, Mansour A. Portal venous gas following colo-noscopy and small bowel follow-through in a patient with Crohn's disease. Endoscopy 2007;39(Suppl. 1):E130.
[12] Tuite DJ, Byrne A, Colhoun E, et al. Pneumatosis intestinalis and portal-venous gas: an unusual presentation of acute appendicitis. Australas Radiol 2007;51:B137—9.
[13] Pan H-B, Huang J-S, YangT-L, et al. Hepatic portal venous gas in ultrasonogram—benign or noxious. Ultrasound Med Biol 2007;33:1179—83.
[14] Tsai C-L, Chen M-J, Tan C-K, et al. Superior mesenteric artery syndrome with hepatic portal venous gas. Med J Aust 2007; 186:48.
[15] Negro U, Verdecchia M, Paci E, et al. Hepatic portal venous gas in a patient with enterovascular fistula. Abdom Imaging 2006;31:706—9.
[16] Nelson AL, Millington TM, Sahani D, et al. Hepatic portal venous gas: the ABCs of management. Arch Surg 2009;144: 575—81. discussion 581.
[17] Lebiedz P, Ullerich H, Seelig M, et al. Jejunal feeding tube causing intestinal pneumatosis and fulminant portal venous gas embolism. Endoscopy 2005;37:926.
[18] Iguchi S, Alchi B, Safar F, et al. Hepatic portal venous gas associated with nonocclusive mesenteric ischemia in a he-modialysis patient. Clin Nephrol 2005;63:310—2.
[19] Lewin M, Pocard M, Caplin S, et al. Benign hepatic portal venous gas following caustic ingestion. Eur Radiol 2002; 12(Suppl. 3):S59—61.
[20] Faberman RS, Mayo-Smith WW. Outcome of 17 patients with portal venous gas detected by CT. Am J Roentgenol 1997;169: 1535—8.
[21] Fataar S, Cadogan E, Spruyt O. Ultrasonography of hepatic portal venous gas due to diverticulitis. Br J Radiol 1986;59: 183—5.
[22] Nachtegaele P, Afschrift M, Vandendriessche M, et al. Sono-graphic diagnosis of gas embolism in the portal vein. Gastro-intest Radiol 1982;7:375—7.
[23] Bodewes HW, Puylaert JB. Ultrasound in detection of portal venous gas in adults. Gastrointest Radiol 1991;16:35—7.
[24] Kinoshita H, Shinozaki M, Tanimura H, et al. Clinical features and management of hepatic portal venous gas: four case reports and cumulative review of the literature. Arch Surg 2001; 136:1410—4.
[25] Lien WC, Chang WT, Huang SP, et al. Hepatic portal venous gas associated with poor outcome in out-of-hospital cardiac arrest patients. Resuscitation 2004;60:303—7.
[26] Lee C-H, Chang Y-C. Mesenteric and hepatic portal venous gas in a woman with pelvic inflammatory disease. J Emerg Med 2012;43:e57—8.
[27] Radin DR, Rosen RS, Halls JM. Acute gastric dilatation: a rare cause of portal venous gas. Am J Roentgenol 1987;148: 279—80.
[28] Chang C-J, Shun H-C, Chuang C-C. Hepatic portal venous gas induced by emphysematous pyelonephritis: a rare case in hemodialytic women. Am J Emerg Med 2009;27:1171. e1—3.
[29] Arbell D, Elazary R, Koplewitz BZ. Renal, pancreatic, splenic, mesenteric, and portal venous gas as a sign of intestinal necrosis. J Pediatr 2010;156. 1031—1031.e2.
[30] Mao Y-C, Wang J-D, Wang L-M. Hepatic portal venous gas caused by emphysematous pyelonephritis. Clin Gastroenterol H 2009;7:A25.
[31] Yamamuro M, Ponsky JL. Hepatic portal venous gas: report of a case. Surg Today 2000;30:647—50.
[32] Naguib N, Mekhail P, Gupta V, et al. Portal venous gas and pneumatosis intestinalis; radiologic signs with wide range of significance in surgery. J Surg Educ 2012;69:47—51.
[33] Hou S-K, Chern C-H, How C-K, et al. Hepatic portal venous gas: clinical significance of computed tomography findings. Am J Emerg Med 2004;22:214—8.
[34] Dill-Macky MJ. Benign hepatic portal venous gas following blunt abdominal trauma. Australas Radiol 1997;41:166—8.
[35] Singh UN, Kapoor A, Aggarwal A, et al. Hepatic portal venous gas and "the aquarium sign" due to intussusception in kawasaki disease. Echocardiography 2012;29:E201—3.
C.-Y. Huang et al.
[36] Hollerweger A, Rettenbacher T. Detection of portal venous gas on sonography, but not on CT. Eur Radiol 2003;13(Suppl. 4): L251-3. author reply L254.
[37] Griffith J, Apostolakos M, Salloum RM. Pneumatosis intestinalis and gas in the portal venous system. J Gastrointest Surg 2006;10:781-2.
[38] Wiesner W, Mortele KJ, Glickman JN, et al. Pneumatosis intestinalis and portomesenteric venous gas in intestinal
ischemia: correlation of CT findings with severity of ischemia and clinical outcome. Am J Roentgenol 2GG1;177:1319—23.
[39] Kernagis LY, Levine MS, Jacobs JE. Pneumatosis intestinalis in patients with ischemia: correlation of CT findings with viability of the bowel. Am J Roentgenol 2003;180:733—6.
[4G] Khorram-Manesh A, Oden A. Management of hepatic portal venous gas and pneumatosis intestinalis in critically sick adult patients. Scand J Gastroenterol 2GG9;44:1G19—2G.
