Skip to main content
Annals of The Royal College of Surgeons of England logoLink to Annals of The Royal College of Surgeons of England
. 2020 May 6;102(7):e173–e175. doi: 10.1308/rcsann.2020.0089

Portal venous gas as a radiological sign in a sigmoid diverticular abscess and its non-surgical management: a case report

CY Kong 1,2,, HL Goh 3, JE Anderson 1
PMCID: PMC7450440  PMID: 32374180

Abstract

A 62-year old man who presented unwell with no specific symptoms or signs was found to have portal venous circulation gas complicating a small diverticular abscess. He was successfully managed with a course of antibiotics and had full resolution of symptoms, therefore avoiding the need for surgical intervention. While most commonly associated with bowel ischaemia and therefore often warranting emergency laparotomy, portal venous gas within the context of other underlying pathology often presents opportunities for delayed surgery or more conservative management options.

Keywords: Portal venous gas, Diverticular abscess, Diverticular disease

Background

Portal venous circulation gas has been traditionally regarded as an ominous radiological sign of intra-abdominal pathology associated with a high risk of mortality and morbidity. In rare instances, this may be secondary to diverticular sepsis, in which invasive surgery may be delayed or avoided.

Case history

A 62-year-old man presented to accident and emergency with a one-week history of feeling generally unwell, reduced appetite and rigours. He had no specific symptoms on systemic enquiry and specifically denied gastrointestinal symptoms including abdominal pain. In terms of past medical history, he had hypertension. He was on regular ramipril, simvastatin and lansoprazole.

Examination was essentially normal except for occasional wheezing on auscultation of the chest. Specifically, he had a soft, non-tender abdomen, with normal bowel sounds.

Initial investigations revealed raised inflammatory markers (C-reactive protein, CRP, 244mg/l and a white cell count of 17.3 × 109/l) and markedly deranged liver function tests (bilirubin 50umol/l, alkaline phosphatase 148U/L, gamma-glutamyltransferase 392u/l, alanine transaminase 174 u/l), and a low platelet count (44 × 109/l). He had normal renal function with a slightly raised urea of 9.9mmol/l and a normal haemoglobin (14.8g/dl). A liver screen, including bloodborne virus screen, was normal.

In terms of an initial septic screen, he had a normal chest x-ray, normal urinalysis. Blood culture results were consistent with an Escherichia coli bacteraemia. The patient was treated for suspected biliary sepsis and started on amoxicillin, metronidazole and gentamicin.

Abdominal ultrasound was reported as showing only non-specific portal tract calcification in a wedge area of the liver.

In view of the non-diagnostic nature of the ultrasound, computed tomography (CT) was performed, which revealed a significant volume of intrahepatic portal gas (Fig 1). Air was present within the inferior mesenteric vein, with extension to the confluence of the portal vein where there was evidence of a developing infective thrombus (Figures 24). Further, the source of the portal venous air was tracked to an intramural abscess measuring 14 × 20mm, secondary to sigmoid diverticulitis with left pericolic stranding (Figures 4 and 5). Related to this were features in keeping with possible liver micro-abscesses. Incidental findings were a cystic area measuring 2.5cm by 2.5cm in the pancreatic tail and a tiny renal calculus in the lower pole of the left kidney (Fig 2). There were no features suggestive of mesenteric infarction.

Figure 1.

Figure 1

Axial view: figure demonstrates multiple streaks of gas extending to the liver periphery consistent with intra-hepatic portal venous gas (red arrow). Note that this differs from pneumobilia (gas within biliary tree) which is of central pattern.

Figure 2.

Figure 2

Axial view: thrombus and gas in the portal vein (red arrow). Incidental pancreatic tail cyst (red asterisk).

Figure 4.

Figure 4

Coronal view: intrahepatic portal venous gas is clearly seen here (red arrows), thrombus and small gas locule within the portal vein (orange arrow). Sigmoid diverticulosis (yellow arrows).

Figure 5.

Figure 5

Axial view: sigmoid diverticular abscess (red arrow).

Figure 3.

Figure 3

Sagittal view: gas tracking up the inferior mesenteric vein (double red arrows).

A decision was made jointly with the patient to trial antibiotics, as abdominal examination was equivocal and in light of the CT findings showing no absolute indications for emergency laparotomy.

Differential diagnosis

The absence of specific signs and symptoms in this patient made initial diagnosis difficult. His deranged liver function tests suggested sepsis from a hepatobiliary source, but the CT revealed the unexpected findings of a diverticular abscess leading to extensive portal venous and intrahepatic gas. Portal venous circulation gas is a radiological sign most commonly associated with mesenteric infarction. His clinical picture and CT findings, including the pattern of air within the venous drainage of the diverticular abscess and absence of radiological signs of bowel ischaemia refuted this diagnosis.

Treatment

The initial antibiotic regime of intravenous amoxicillin, metronidazole and gentamicin was associated with a good clinical response initially. However, the patient continued to be intermittently pyrexial and his inflammatory markers remained high. In light of this and updated microbiology results, the patient’s antibiotics were changed to intravenous piperacillin-tazobactam, metronidazole and ciprofloxacin.

He made a good clinical and radiological response, with an interval CT showing resolution of portal venous gas and the diverticular abscess. He was discharged home but continued on outpatient antibiotic therapy (oral ceftriaxone, metronidazole and ciprofloxacin) for a total of six weeks. He was also anticoagulated with apixaban for his portal venous thrombosis.

Outcome and follow-up

He continued to be reviewed by the infectious diseases team and had normal CRP and white cell count at the end of his course of antibiotics. An outpatient colonoscopy confirmed severe sigmoid diverticular disease but no other pathology. As he remained asymptomatic and well, the decision was made to not pursue resection of the diverticular segment. At one-year follow-up, he remains well and asymptomatic.

Discussion

Portal venous circulation gas is well regarded as a radiological sign associated with severe intra-abdominal pathology with a concomitant high risk of mortality and morbidity. In adult patients, the most common underlying pathology to cause this sign remains bowel ischaemia and/or infarction.1 Portal venous circulation gas is, however, not pathognomonic of bowel ischaemia. Only 60–70% of patients exhibiting this sign are found to have underlying bowel ischaemia.1,2 Other causes of portal venous circulation gas include sepsis, bowel obstruction, cancer, inflammatory conditions and trauma.1,2

There remains a paucity in studies elucidating the pathophysiology of portal venous gas, although popular hypotheses include translocated bowel luminal air and gas-producing pathogens.2

Early reviews of the outcomes of patients with this sign revealed a mortality in excess of 70%.2 More recent studies have shown a halving of mortality, which has been hypothesised to be related to CT picking up more benign associated conditions.2

Portal venous circulation gas remains a rare sign manifesting in diverticular disease. To date, there are only a couple of contemporary case reports describing this association, which reported therapeutic success with delayed surgery, therefore suggesting that portal venous gas on CT, within the context of diverticular sepsis, should not be viewed as an absolute indication for emergency surgery, as it is in bowel ischaemia.3,4 Our case provides an example where no surgical intervention was required and suggests that medical therapy and careful follow-up may be effective in this group of patients.

The utility of this sign as a marker of severity and prognosis in bowel ischaemia remains uncertain, and in the presence of other underlying pathology, this uncertainty is magnified.2

Conclusions

Portal venous gas, while regarded as an ominous sign, is not specific to mesenteric infarction. In certain underlying pathology, for example diverticular sepsis, antibiotic treatment and careful observation should be considered a possible option, allowing the delay or avoidance of an exploratory laparotomy. This may be useful in the context of patients unfit for surgery, who would otherwise be palliated.

References

  • 1.Abboud B, El Hachem J, Yazbeck T et al. Hepatic portal venous gas: physiopathology, etiology, prognosis and treatment. World J Gastroenterol 2009; : 3585–3590. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Nelson AL, Millington TM, Shani D et al. Hepatic portal venous gas: the ABCs of management. Arch Surg 2009; : 575–581. [DOI] [PubMed] [Google Scholar]
  • 3.Nobili C, Uggeri F, Romano F et al. Pylephlebitis and mesenteric thrombophlebitis in sigmoid diverticulitis: medical approach, delayed surgery. Dig Liver Dis 2007; : 1088–1090. [DOI] [PubMed] [Google Scholar]
  • 4.Zielke A, Hasse C, Nies C et al. Hepatic-portal venous gas in acute colonic diverticulitis. Surg Endosc 1998; : 278–280. [DOI] [PubMed] [Google Scholar]

Articles from Annals of The Royal College of Surgeons of England are provided here courtesy of The Royal College of Surgeons of England

RESOURCES