Atypical presentation of Lemierre syndrome in a young healthy man with acute jaundice
- James Patrick King ,
- Lauren Matthews and
- Shanze Ashai
- Jack Steinberg Critical Care Unit, King's College Hospital, London, UK
- Correspondence to Dr James Patrick King; james.king16@nhs.net
Abstract
This report presents a case of Lemierre syndrome caused by Fusobacterium necrophorum in a healthy young adult who presented atypically with shortness of breath and jaundice but no clinical or diagnostic evidence of thrombophlebitis. Due to this unusual presentation with jaundice, diagnosis was challenging and delayed. However, the patient was successfully initiated on a prolonged course of intravenous antibiotics; he required a period in the intensive care unit and was discharged without significant complications. This report aims to raise awareness of the diagnosis and treatment of this rare condition and to highlight both common and unusual presentations of the syndrome.
Background
In 1936, André Lemierre described a series of 20 cases of septicaemias arising from inflammatory lesions of the oropharynx/nasopharynx giving rise to distant septic emboli, likely spread via thrombophlebitis of the internal jugular vein (IJV).1 The most common organism responsible was subsequently identified as Fusobacterium necrophorum, 2 and recent studies have suggested that the eponymous ‘Lemierre syndrome’ is rare with an incidence of around five cases per million each year.3 Lemierre syndrome most often develops in young and healthy patients.4 It is not well understood how this bacterium occasionally invades the lateral pharyngeal space to enter the IJV and blood stream. This vascular invasion has been hypothesised to be related to specific pathogenic properties of Fusobacterium spp, allowing the species to both potentiate and use endothelial dysfunction created by a local inflammatory response.2 5 The majority of patients also develop metastatic septic emboli, most commonly seen within the lungs.6–8
Clinical presentations are varied and include fever (92%–98%), dyspnoea (62%), cough (31%), clinical pneumonia (49%–54%), sepsis on admission (83%) and neck swelling (42%).3 4 The incidence of head/neck vein thrombosis at diagnosis has been suggested to be 80% in a large recent meta-analysis.9 Jaundice has also been mentioned to be a feature of presentation in some, and it has been suggested that the toxic effects of F. necrophorum endotoxins may be implicated in liver dysfunction.7 Lemierre syndrome (LS) is primarily a clinical diagnosis, confirmed by positive blood culture and backed up by evidence of septic emboli or thrombophlebitis.6 7 10 11
Treatment includes empiric prolonged antibiotic therapy, often with a carbapenem, broad-spectrum penicillin and/or metronidazole, guided by sensitivities if available.10 11 Treatment may include anticoagulation, as the condition poses a risk of thrombosis, although the benefits of this are still debated.2 12 As Lemierre syndrome is a rare condition, diagnosis is often delayed and most clinicians, understandably, lack experience in treating this condition. While diagnosis is even more challenging in patients presenting with atypical features, it remains possible if the value of key findings is remembered, such as a history of recent oropharyngeal infection and the finding of septic pulmonary emboli.
Case presentation
A previously fit and well man in his 30s presented to the emergency department (ED) with a 24 hour history of sore throat, cough and lethargy and was prescribed a 10-day course of penicillin-V for presumed bacterial tonsillitis due to clinical findings of enlarged, erythematous tonsils with exudate. He re-presented 7 days later via ambulance with shortness of breath and fever, which had worsened over the interim, and new onset jaundice. On assessment in the resuscitation area of ED at second presentation, he was febrile at 38°C, tachypnoeic with a respiratory rate of 40 breaths per minute, tachycardic with a heart rate of 140 beats per minute, hypotensive with a blood pressure of 85/40 mm Hg and had an initial oxygen requirement of 60% fractional inspired oxygen delivered via venturi mask. On clinical examination, the main finding was jaundice with the tonsils not described as still enlarged or exudative, with good dentition but a single cracked tooth. He was treated for sepsis of unknown origin with piperacillin/tazobactam (Tazocin) and aggressive intravenous fluids. Extensive microbiological and serological investigations were ordered.
Investigations
The patient’s initial chest X-ray showed patchy airspace opacification in the left lower lobe, periphery of the left lung and right upper lobe (left, figure 1). For comparison, a repeat X-ray 7 days into admission is included on the right hand side of figure 1.
Admission CXR (left) versus repeat CXR 7 days postadmission.
Initial blood results revealed markedly raised inflammatory markers, with a total white cell count of 16.75 × 109/L, neutrophils of 15.71 × 109/L, C reactive protein of 280 mg/L and ferritin 6766 µg/L. Renal function initially showed an acute kidney injury (AKI) with a creatinine of 258 µmol/L and an eGFR of 57 mL/min; this corrected with fluid resuscitation. Initial liver function tests were deranged showing a bilirubin level 216 µmol/L, alkaline phosphatase 215 IU/L, aspartate transaminase 686 IU/L and gamma-glutamyl transferase of 59 IU/L. An acute hepatitis liver screen did not reveal obvious cause for this derangement. Viral screen for HIV 1&2, hepatitis A/B/C, adenovirus, CMV, EBV and a viral respiratory screen were negative. He had not been vaccinated against COVID-19 but repeatedly swabbed negative over the course of his admission. Urinary legionella and pneumococcal antigens were negative, as were ASOT, leptospira and treponemal serology. Immunoglobulin screen, ANA, anti-MPO and anti-PR3 were also negative. Blood cultures taken in ED prior to antibiotics grew F. necrophorum, which was relayed to the intensive care unit (ICU) team on day 3 of admission.
Prior to the blood culture result, to identify a source of sepsis and establish if a structural cause for the acute hepatitis picture was present, he underwent a CT abdomen-pelvis, which showed normal appearances of the liver and other intra-abdominal viscera. A triple-phase liver CT was also normal, though detected consolidation at the lung bases and prompted specific lung imaging. A CT-Pulmonary Angiogram showed bilateral small pleural effusions associated with posterior lower lobe relaxation atelectasis and, in addition, multiple peripheral lung nodules, some of which were cavitating.
Following isolation of the causative microbe, a diagnosis of Lemierre syndrome was considered, and a CT neck was performed to image the IJV. This was reported with no collections or thrombosis but an appearance suggesting uncomplicated inflammation on the right side of the oropharynx in keeping with pharyngitis. A repeat CT thorax, abdomen and pelvis were performed later in hospitalisation and showed an unchanged appearance of the lung nodules with an increase in bilateral pleural effusions, with findings in keeping with septic emboli. A section of this is demonstrated below in figure 2.
Axial slice of CT thorax showing bilateral pleural effusions and atelectasis with septic lung nodules.
Furthermore, given the patient’s previous reported cracked tooth in the context of Lemierre syndrome, a maxillofacial surgical review was requested. The maxillofacial team felt there were not clinical signs of infection/cysts around the cracked tooth and that it was felt to be unlikely to be contributing to his Lemierre syndrome.
Treatment
While these investigations were ongoing, the patient was treated in the ICU with norepinephrine, which he required in total for 5 days, intravenous fluids and Tazocin. Following the discovery of F. necrophorum in blood culture, on the advice of microbiology specialist review and antimicrobial sensitivity testing, metronidazole was added. The repeat CT thorax findings of increasing bilateral pleural effusions prompted the cardiothoracic team to be contacted and place bilateral surgical chest drains within the Intensive Care Unit, which in total drained over 2 L of bloody/serous fluid. This fluid was also sent for Microscopy, culture and sensitivity culture which was negative. These drains were removed after 48 hours, and his oxygen requirements reduced to room air. Given his clinical improvement, the patient was stepped down from a level 3 setting to a medical ward where he continued his long course of antibiotics guided by regular contact with the microbiology team. Of note, during admission, he initially received daily prophylactic dosing of unfractionated heparin, which was then switched to low molecular weight heparin after renal function improved.
Outcome and follow-up
After discharge from the ICU, the patient remained clinically stable. His inflammatory markers decreased, kidney and liver function normalised and his jaundice improved. Accordingly, intravenous antibiotics were switched to oral amoxicillin-clavulanic acid with metronidazole for a total of 6 weeks. After discharge, the patient was followed up every 2 weeks by the infectious disease team. At his initial telephonic review 2 weeks after hospital discharge, he described an ongoing cough (non-productive) and shortness of breath. In a later outpatient clinic, he mentioned still feeling very fatigued. He had no further fevers or sweats, as he initially had on leaving hospital. He also mentioned the pain at the sites of his chest drains, and internal jugular central line was steadily improving, as was his mobility. His infectious markers had resolved by this point.
Discussion
Since Lemierre’s original description, there has been a lack of literature consensus regarding which infective clinical features constitute a classical diagnosis of ‘Lemierre syndrome’ and which infective agent is causative.10 11 It has been suggested that LS is used as a catchall for all cases of F. necrophorum bacteraemia with primary foci in the head;7 however, most studies prefer more specific definitions, which may include a head-neck infection, thrombosis of a local vein (most commonly, but not exclusively, the IJV) and/or evidence of peripheral septic embolisation.8 11 An extensive review collating current case series of LS defines LS with the following criteria: (1) preceding oropharyngeal infection, (2) evidence of metastatic infection and (3) evidence of IJV thrombophlebitis or isolation of Fusobacterium on blood cultures.11
While Lemierre himself suggested that it was ‘relatively easy to make a diagnosis on the simple clinical findings…mistake is almost impossible’,1 10 it has since been touted as a ‘forgotten disease’4 10 with a rare but rising incidence.3 4 10 Using the above consensus criteria, our patient had a confirmed F. necrophorum bacteraemia, a radiologically identified pharyngeal focus of infection, and multiple septic pulmonary metastases which fits the definition of previously defined cases of Lemierre syndrome.
This case is however notable in two aspects. First, the lack of clinical or radiological evidence of IJV thrombophlebitis, which is thought to be a crucial step for bacteraemia and subsequent metastatic septic emboli.1 3 7 13 Retrospective studies of 24 patients with Lemierre syndrome in Denmark found 42% patients presented with neck swelling,4 while a similar retrospective study of 104 Swedish patients found 70% had evidence of thrombosis, the majority of which were IJV thrombi.3 A recent meta-analysis of 712 Lemierre syndrome patients suggested that 83.8% of patients had an initial thrombosis of a head/neck vein.9 Our patient did not complain of neck pain or swelling during or on admission, and CT neck did not show radiological evidence of IJV thrombophlebitis, although other features were in fitting with the syndrome. A secondary review of the patient’s scans by a specialist base-of-skull and neck radiologist confirmed that there was no radiological evidence of thrombosis or thrombophlebitis. However, while IJV thrombosis is a common and important feature of LS, it is not felt to be a standalone diagnostic criterion.8 11 Instead, the presence of pulmonary septic emboli has been suggested as a surrogate marker that thromboses have already metastasised,11 and one study found that while 62% of 82 LS patients had radiological evidence of IJV thrombosis, 32% of those without IJV thrombosis still showed evidence of septic embolisation to the lungs.12
Second, the patient was unusual in his presentation with significantly deranged liver function tests and frank jaundice. In the aforementioned Swedish study, the average bilirubin on admission was 19.5 µmol/L (IQR 15–35 µmol/L), compared with 216 µmol/L for our patient.3 Despite the patient having had a recently resolved oropharyngeal infection, the deranged liver function was of significance as it became the focus of investigations for the initial hours and days after presentation, delaying the diagnosis or consideration of oropharyngeal infection and Lemierre syndrome. In addition, the lack of evidence for clinical thrombophlebitis was important as, until isolation of the organism, this did not prompt clinicians to consider the diagnosis. It is interesting to note that Lemierre’s original paper suggested that icterus and subicterus were ‘invariably present’ in his subjects.1 In patients presenting with impaired liver function, the possibility of hepatic septic lesions should be considered, which have been reported in 2% of the patients.9 However, no liver abscess was found in our patient, suggesting that liver impairment may have been associated with sepsis
Two similar case reports presenting with jaundice and AKI were identified in the literature. The first case outlined a previously healthy male in his twenties who presented jaundice with sore throat, fever, diarrhoea and AKI. With this case, it was not until neck pain was noted that Lemierre syndrome considered and diagnostic investigations sent.14 This patient also improved with oxygen support, intravenous antibiotics, High Dependency Unit admission and intravenous fluids. A second case in a previously healthy male in his thirties presented with a similar clinical picture, but he was initially worked up for leptospirosis.15 This second patient did not show any clinical or ultrasound evidence of IJV thrombosis but also improved with intravenous antibiotics, HDU admission and haemodialysis.
In conclusion, Lemierre syndrome should be considered in any young otherwise healthy patient presenting with recent or current oropharyngeal infection, especially in the presence of septic pulmonary emboli, even in the absence of neck findings. This can be considered therefore prior to the microbiological diagnosis of F. necrophorum and in the presence of atypical features such as jaundice and deranged liver function tests.
Patient’s perspective
I felt I received a diagnosis quite promptly and that it was explained well, considering it is a rare condition. I did not find it particularly difficult or upsetting that I had a rare condition and did not have particular complaints or comment about the care while on ICU or the wards. However, I found the chest drains quite painful and they were the most difficult part of my stay and was also frustrated by the length of time I had to spend in hospital. In addition, I am still bothered by pain and fatigue, but feel this is steadily improving. I have been seen in clinic as an outpatient with repeat investigations and monitoring of my condition by the Infectious Diseases Team.
Learning points
-
Lemierre syndrome is a rare condition defined by Fusobacterium necrophorum bacteraemia originating from an infective source in the head or neck, typically an oropharyngeal infection, with classical complications of internal jugular vein (IJV) thrombophlebitis and metastatic septic emboli (commonly in the lungs).
-
Atypical complications may also atypically include jaundice and deranged liver function tests.
-
Not all cases of Lemierre syndrome appear to present with clear or ongoing neck symptoms, and there may not be diagnostic evidence of IJV thrombophlebitis.
-
The syndrome can cause severe sepsis, and patient’s may require high-dependency or intensive care level support.
-
Treatment involves a long course of antibiotics guided by sensitivities, often a broad-spectrum penicillin or carbapenem with metronidazole.
Ethics statements
Patient consent for publication
Footnotes
-
Contributors All authors were directly involved in the care of the patient. JPK and LM were involved in the drafting and acquisition/analysis of data. LM was also involved in correspondence with the patient. SA was involved in drafting and correspondence with relevent specialties.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
-
Competing interests None declared.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2022. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer