Septic Thrombophlebitis in the Portal Veins: A Case of Pylephlebitis Linked to Colo-Venous Fistula and Diverticulitis

Introduction

Diverticulitis occurs in 10–25% of patients with diverticulosis and is considered a common disease in the Western world [1]. The Western standard of living, such as a low-fiber and high-fat diet, high meat consumption, obesity, and the use of certain drugs promotes the development of diverticulitis [1]. The symptoms mainly include left-sided lower abdominal pain and stool irregularities [1]. The criterion standard of diagnosis is contrast-enhanced CT [1]. Depending on its severity, diverticulitis is treated conservatively with antibiotics and dietary restrictions, or surgically [1]. In some cases, complications occur, such as intra-abdominal abscesses, fistulas, intestinal obstruction, frank perforation [1] or pylephlebitis, which is a rare complication [2–5]. Falkowski states correctly: “Although diverticulitis is the most common cause of pylephlebitis, pylephlebitis is a rare complication of diverticulitis” [5].

Pylephlebitis is a septic thrombosis of the portal vein or one of its tributaries and usually occurs as a rare complication of septic disease in the portal venous drainage area [2–4]. In a systematic review from 2023, diverticulitis (26.5%) and acute appendicitis (20%) were found to be the most common causes of pylephlebitis [6]. The annual incidence ranges from 0.37 to 2.7 per 100 000 inhabitants [6]. Pylephlebitis is associated with high morbidity and mortality [4,6]. In recent case studies, mortality rates were 11–32% [4,7]. The main concerns are uncontrolled spread of the infection via the vascular system, liver abscesses, and septic embolisms with consecutive infarction [2,5,8]. Plemmons et al therefore conclude that pylephlebitis is a rare but dramatic complication [2]. It is caused by a polymicrobial infection mostly with E. coli, Bacteroides species (spp.), and Streptococcus spp. [4,6]. The symptoms are non-specific [3,4,9]. The most common are fever, abdominal pain, nausea, and diarrhea [6,9,10]. The criterion standard of diagnosis is contrast-enhanced CT [3,9,11]. In addition to thrombosis, CT also frequently shows air in the portal venous vessels as a sign of ischemia [3,12]. The treatment involves an antibiotic regimen narrowed according to the resistogram [10]. Due to a lack of data, there is no consensus on the role of anticoagulation concerning the treatment of pylephlebitis [10]. This report describes the case of a 43-year-old man with pylephlebitis, or septic thrombophlebitis of the portal veins, associated with colo-venous fistula and sigmoid diverticulitis.

Case Report

A 43-year-old man presented to the emergency department after collapsing at home. The previous week, he had received antibiotic treatment at an out-of-town hospital for an infection of unknown focus. After discharge, he had unspecific symptoms, including cephalgia, lumbago, diffuse abdominal pain, and palpitations. His history was significant for hypertension, type II diabetes mellitus, obesity, and alcohol abuse. Apart from a laparoscopic cholecystectomy, there were no previous operations. He was not taking any medication and had no significant family history.

On physical examination, he was tachycardic (140 beats/minute), with chills and fever up to 38.3°C. The remaining examination findings were unremarkable, and the abdomen in particular was soft and without tenderness. Initial laboratory investigations were significant for an infection, with a C-reactive protein level of 229 mg/L, leukocytosis at 31 leukocytes/nL, and procalcitonin level of 136 µg/L. The lactate level was within normal range. Summarizing the results of the physical and laboratory examination, the patient presented as septic. A series of further diagnostics followed to identify the focus of the septic event. For example, urine analysis by dipstick which showed no diagnostic abnormalities, and a chest radiograph and an abdominal sonography were also unremarkable. CT with contrast of the abdomen and pelvis finally showed a sigmoid diverticulitis with covered perforation, forming a colo-venous fistula to the superior mesenteric vein, with gas trapped in the hepatic portal branches and a pylephlebitis of the superior mesenteric vein (Figures 1, 2).

After the diagnosis was made, the one-hour sepsis bundle was immediately completed by calculated intravenous antibiotic therapy with ceftriaxone and metronidazole, as well as intravenous volume substitution. An emergency laparotomy was indicated and the patient was rushed to the operating theatre. Intraoperatively, the sigmoid colon was found to be perforated with adherent small bowel and abscessation into the mesenteric root. The pylephlebitis of the superior mesenteric vein was clearly palpable. An en bloc resection of the sigmoid and segmental resection of the adherent ileum loop were performed. As the patient was young and had no signs of a generalized peritonitis, a colo-rectostomy and an ileo-ileostomy were created. Considering the large size of the septic thrombus of the vena mesenterica superior and the high risk of embolization, a thrombectomy was not performed.

After surgery, the patient was transferred to the normal ward. Intravenous antibiotic therapy was initially continued postoperatively with ceftriaxone and metronidazole. When blood cultures were reported to be positive for Proteus mirabilis, E. coli, and Enterococcus durens, the antibiotic treatment was continued intravenously with vancomycin and meropenem according to the resistogram. In addition to treating the primary focus of the infection, antibiotic therapy was also used to combat pyle-phlebitis of the superior mesenteric vein. The pylephlebitis was also treated with anticoagulation. Postoperatively, the patient was anticoagulated with low-molecular-weight heparin. Due to a wound healing disorder in the further postoperative course, the inpatient stay was prolonged. The wound was partially opened and healed secondarily. The patient was finally discharged on the 19th postoperative day in good health. The antibiotic treatment and the anticoagulation were continued after the inpatient stay to resolve the pylephlebitis. After discharge, the antibiotic regimen was continued orally with cefpodoxime, using its vast first-pass effect, which guarantees maximum effect in the area of the thrombus, without systemic adverse effects. The low bioavailability of cefpodoxime was therefore not an obstacle, but a decisive advantage. Antibiotic treatment was given for a total of 4 weeks postoperatively. After discharge, an anticoagulation with rivaroxaban was established for 6 months. A CT scan taken 8 weeks postoperatively showed no further evidence of the septic thrombus in the superior mesenteric vein.

Discussion

Case reports such as this one are intended to create a better understanding of the rare clinical condition of pylephlebitis, gathering information about cause, presentation, diagnosis, treatment, and their pitfalls.

Pylephlebitis typically occurs in the context of an intra-abdominal infection or abdominal sepsis [2,3,11]. The diagnostic criterion standard is CT, as it can identify the thrombus itself as well as the source of infection [6]. Due to the unspecific symptoms, CT is often not performed upon admission, often delaying the diagnosis of pylephlebitis [6,8]. The time from initial symptoms to diagnosis therefore varies from days to several weeks [6,11,13,14]. Often, only the occurrence of an acute abdomen leads to extended diagnostics using CT [6]. Even in our severe case with the development of a colo-venous fistula, 12 days passed between the first symptoms and the patient being diagnosed. In similar cases by Guerra et al and Youssef et al, 2 weeks passed before the diagnosis was established [3,8].

CT almost always shows filling defects in the superior mesenteric vein or its tributaries [15]. The presence of air trapped in the superior mesenteric vein, in the portal vein, or in the liver may then be an expression of possible intestinal ischemia [16,17], or, as in our case, may be due to a colo-venous fistula. Falkowski et al and Youssef et al however describe a case in which intraluminal gas in the portal venous system was the sole expression of suppurative thrombophlebitis, as neither ischemia nor a fistula was present [3,5]. Due to the high mortality and fatal complications of pylephlebitis, patients should receive treatment immediately after the diagnosis has been established [8].

The treatment includes surgical removal of the center of inflammation and initial intravenous and later oral antibiotic therapy [6,9]. The need for surgical treatment is undisputed due to the severity of the disease. The few similar cases in which a conservative approach was initially pursued either failed or led to long stays in intensive care units [3,5]. Antibiotics should also be used in all patients [11]. Initially, a broad-spectrum antibiotic is used, which is adjusted according to the resistogram over the course of treatment [6,10]. Most cases are treated with antibiotics for 4–6 weeks. [6,18]. To date, no randomized controlled trials have been conducted to evaluate empirical antibiotic regimens or the use of anticoagulants. While antibiotic treatment of pylephlebitis is undisputed, the benefit of anticoagulation has long been the subject of controversy due to a lack of data [5,18]. However, more recent and larger case series demonstrate the benefit of anticoagulants in the treatment of pylephlebitis [4,7,19].

The aim of anticoagulation is to prevent the progression of thrombosis or its migration, to maintain vascular patency, or achieve recanalization and thus avert the development of portal hypertension, embolisms, and infarctions [20,21].

Studies have shown that the recanalization rate of the portal vein is increased by the administration of anticoagulants (58% vs 21%) and at the same time anticoagulants decrease the risk of developing portal hypertension [4,19,22]. In their meta-analysis of 100 individual case reports published between 1971 and 2009, Kanellopoulo et al showed that anticoagulation is associated with lower mortality compared to antibiotic therapy alone [22]. The unrestricted use of anticoagulants is countered by the view that anticoagulation should only be established under certain conditions. For example, if the imaging shows progression of the thrombus or if the patient presents with persistent fever, cancer, thrombophilia, or signs of portal hypertension or intestinal ischemia [6,20].

In the present case, we opted for anticoagulation due to the extensive thrombosis and the concern about additional intestinal ischemia if the thrombus increased. As no consensus exists on the optimal length of time to administer anticoagulation, we based our decision on the recommended duration of anticoagulation in patients with deep venous thrombosis and pulmonary artery embolism.

There are few reports on the success of more invasive therapeutic methods such as thrombectomy. Nishimori et al removed a septic thrombus of the portal vein and the superior mesenteric vein in the context of acute appendicitis using a Fogarty catheter. The removal of the thrombus is undisputedly of paramount importance to prevent complications such as septic embolisms, liver abscesses, infarctions, or portal hypertension.

Currently however, treatment options such as catheter lysis or surgical thrombectomy cannot be recommended as first-line therapy due to insufficient data, questionable effectiveness, the risk of re-thrombosis, and a high procedure-related complication rate [23,24]. Although smaller case series with interventional thrombosis have shown promising outcomes, the results are generally inhomogeneous [25–28].

Conclusions

Pylephlebitis is a rare and potentially lethal complication of an intra-abdominal septic inflammatory process. In everyday clinical practice, the relatively unspecific clinical characteristics make early diagnosis difficult and do not immediately suggest pylephlebitis. Antibiotic therapy appropriate to the resistogram for at least 4 weeks and treatment with anticoagulants form the cornerstones of initial therapy, in addition to surgical removal of the inflammatory process. The role and unrestricted use of anticoagulation in pylephlebitis continues to be the subject of controversy due to the unclear data, and is decided on an individual basis.