Skip to main content
BMJ Case Reports logoLink to BMJ Case Reports
. 2020 Dec 31;13(12):e237061. doi: 10.1136/bcr-2020-237061

Hafnia alvei pneumonia: a rare cause of infection in the multimorbid or immunocompromised

Austin Begbey 1, Joseph Henry Guppy 1,✉, Chithra Mohan 1, Simon Webster 1
PMCID: PMC7780556  PMID: 33384344

Abstract

Hafnia alvei is a rare, poorly understood commensal bacterium which has, on occasion, been shown to infect humans. We present two cases. The first patient presented with a 1-week history of dyspnoea, pleurisy and a productive cough, and the second with a prodrome of fatigue and night sweats. The former had a history of severe chronic obstructive pulmonary disease and the latter had a history of Crohn’s disease. Both patients had underlying comorbidities and immunosuppression, but differed in presentation, radiological findings and recovery. This case series aims to remind readers of the broad differential of pathogens that can lead to disease in the immunocompromised and that one should not dismiss atypical cultured bacteria as commensal too hastily.

Keywords: infections, respiratory system, pneumonia (respiratory medicine), pneumonia (infectious disease)

Background

Hafnia alvei is a Gram-negative, facultative anaerobic bacillus of the Enterobacteriacae family.1 Its role in human infection is poorly understood, with few studies investigating this bacterium, most of which were published more than three decades ago.1–4

H. alvei pneumonia is a rare diagnosis. Nevertheless, over the course of 3 months, two patients presented to our district general hospital with pneumonia and H. alvei isolated from their respiratory tracts. The cases expand on the limited publications surrounding this bacterium and strengthen the evidence to demonstrate which subgroups of patients may be at risk of developing this infection and its manifestations.

Case presentation

Patient 1

A 73-year-old woman with known chronic obstructive pulmonary disease (COPD) presented to the accident and emergency department following a 5-day history of increasing shortness of breath at rest, productive cough of white sputum and a 1-day history of severe right-sided pleuritic chest pain. She had taken 7 days of oral prednisolone. Her COPD was severe with a forced expiratory volume in 1 s (FEV1) 20% of predicted (0.43 L) and FEV1/forced vital capacity of 35% (0.43/1.6 L). She had suffered two exacerbations within the last year, neither of which required hospital admission. She was an ex-smoker from 2012, with a 55 pack-year history.

Other medical history included paroxysmal atrial fibrillation, hypertension, dyslipidaemia, impaired glucose tolerance and transient ischaemic attack. Her medications included warfarin, a tiotropium bromide inhaler, a budesonide and formoterol combination inhaler, and phyllocontin. She lived independently with her husband.

On examination in the accident and emergency department, chest auscultation revealed right basal crepitations and decreased breath sounds bilaterally. She was saturating at 92% on 1 L via nasal cannulae, with a respiratory rate of 16/min. She had a regular heart rate of 102/min and blood pressure was 92/47 mm Hg. Her jugular venous pressure was not elevated and her heart sounds were normal. She had a mildly tender right costal margin with a soft abdomen. She remained without fever throughout.

Investigations

Her ECG confirmed sinus tachycardia, with left axis deviation. Her erect chest radiograph showed right basal infiltrates in keeping with a right lower lobe pneumonia. This was confirmed on CT which showed solid consolidation at the posterior right lower lobe, with bilateral apical fibrosis and severe emphysematous changes throughout both lung fields (figures 1 and 2).

Figure 1.

Figure 1

Erect chest radiograph of patient 1, right lower zone opacity in keeping with bronchopneumonia. Note the left upper zone patchy opacities were long-standing.

Figure 2.

Figure 2

CT scan of patient 1, dense consolidation in the right lower lobe.

Her blood tests on presentation were largely unremarkable except for a mild neutrophilia. Her arterial blood gas on 1 L oxygen showed type 1 respiratory failure with a PaO2 of 7.9 kPa.

Differential diagnosis (pre-imaging)

The most likely differential lay between an infective exacerbation of COPD and community-acquired pneumonia. Chest X-ray was required to clarify this. The use of corticosteroids has been linked with an increased risk of developing pneumonia.5

Given her acute presentation of dyspnoea, hypoxia and cough, pneumothorax or pulmonary embolism was possible. The incidence of secondary pneumothorax with COPD is 26 per 100 000, and the risk increases with the severity of COPD.5 Given her anticoagulated state with warfarin, pulmonary embolism was unlikely, but not impossible.6 In one of the first randomised controlled studies exploring the use of warfarin in atrial fibrillation, the Atrial Fibrillation, Aspirin and Anticoagulation study, 1.5% of patients receiving warfarin sustained a thromboembolic event (n=335).7

Treatment

She was diagnosed with community-acquired pneumonia, with a calculated with a calculated CURB-65 score (confusion, urea, respiratory rate, blood pressure, age over 65) of 2, for his age and low diastolic blood pressure. A sputum sample was sent for microscopy, culture and sensitivity, and she was treated with empirical antibiotics (co-amoxiclav and clarithromycin), nebulised saline and bronchodilators. The sputum culture subsequently grew H. alvei, resistant to penicillins. After being prescribed ceftazidime according to sensitivities, she improved drastically and was discharged home after 5 days.

Outcome and follow-up

She returned to hospital 2 months later with non-specific symptoms of an exacerbation of COPD requiring less than 24 hours admission.

Case presentation

Patient 2

A 20-year-old man with Crohn’s disease, maintained on azathioprine and adalimumab, was admitted with a 4-week history of fatigue and night sweats. He had been recently weaned off corticosteroids. On examination, he was febrile at 38.6°C, had bibasal crackles and normal heart sounds on auscultation.

Investigations

Laboratory investigations showed a white cell count within the normal range and a raised C reactive protein of 46. His liver and renal function tests were also within normal limits. Erect chest radiograph showed new widespread bilateral infiltrates. CT chest showed widespread patchy and multifocal ground glass and ‘tree-in-bud’ opacities, predominantly in the upper and lower lobes (figures 3 and 4).

Figure 3.

Figure 3

Erect chest radiograph of patient 2, multifocal bilateral opacities mostly in upper and middle zones.

Figure 4.

Figure 4

CT scan of patient 2, ‘tree-in-bud’ appearance, particularly in the right lower lobe.

Differential diagnosis

This patient’s differential was more complex, given the CT findings. ‘Tree-in-bud’ changes occur in a wide variety of disease processes. ‘Tree-in-bud’ changes have been well described in Mycobacterium species infection such as tuberculosis and Mycobacterium avium. As the patient was immunosuppressed, these were important diagnoses to exclude,8–10 but were less likely as the patient had been screened for latent tuberculosis prior to starting his immunotherapy. His immunosuppression would also increase his risk of bacterial, fungal and cytomegalovirus infections, all of which can present with non-specific CT findings, such as ‘tree-in-bud’.11

Inflammatory bowel disease can be associated with bronchiectasis and12 13 significant mucous plugging can manifest a similar CT. This could also explain his limited clinical improvement with several short courses of antibiotics.

Chronic disease processes such as obliterative bronchiolitis can be associated with inflammatory bowel disease and present with the ‘tree-in-bud’ changes; however, they typically present with slow, insidious breathlessness, rather than with the symptoms of acute infection our patient exhibited.14

Treatment

A fiberoptic bronchoscopy was performed and bronchial washings grew H. alvei and Serratia marcescens; no acid-fast bacilli were seen or subsequently cultured after 42 days. The patient was initially prescribed intravenous co-amoxiclav, but changed to oral ciprofloxacin as per sensitivities, for a 7-day course. The patient was clinically well despite widespread pneumonic changes and was discharged after 4 days of hospital stay.

Outcome and follow-up

Unfortunately, on cessation of treatment, his symptoms relapsed with persistent fever and cough. He received a further 7-day course of co-trimoxazole, in accordance with the initial sensitivities.

Discussion

H. alvei is very rarely thought to be pathogenic. Extraintestinal invasive infections caused by H. alvei usually occur in patients with chronic diseases or immunosuppression.1

H. alvei is most often found as an enteric commensal, and in cases of immunosuppression, it is also found in the oropharynx.1 2 The name is derived from the old name for the city of Copenhagen, ‘Havn’, and the Latin for beehive, ‘alveus’.2 It was first considered a unique subgroup of enteric bacteria when first described in Møller’s ‘Distribution of amino acid decarboxylases in Enterobacteriaceae’ in 1954.3 It is morphologically and phenotypically similar to the Klebsiella tribe.15 Most of the available literature notes that it rarely infects Homo sapiens although there have been multiple outbreaks among poultry, fish, horses and goats.5 16 17 In the rare instances it does infect H. sapiens, it most commonly causes gastroenteritis, septicaemia of enteric or urinary origin, or bronchopneumonia.18 19

H. alvei has been shown to be sensitive to aminoglyclosides, cephalosporins, carbapenems, quinolones, nitrofurantoin and folate pathway inhibitors. They have resistance to tetracyclines, amoxicillin, ampicillin, azithromycin and fosfomycin.20

Despite the dearth of literature describing H. alvei, interestingly, there were consistencies between our cases described above and those described elsewhere. Most notably, both patients fit within the demographics noted in Gunthard and Pennekamp’s paper, that is, those of multimorbidity or immunosuppression.19 They found that in a sample of 80 H. alvei isolates from 61 patients, 57 patients had underlying health conditions such as haematological, abdominal or pulmonary malignancy, or post-surgery. Five per cent of patients had a history of COPD. It was most often found in a mixed growth alongside other enterobacters, yeast, streptococci or staphylococci. This paper did not comment on the dose of corticosteroids the patients with COPD were taking but based on the theory of immunosuppression leading to increased virulence, it would be interesting to explore the dose–response relationship between H. alvei and immunomodulatory therapies. The use of inhaled corticosteroids in COPD has been repeatedly shown to increase the risk of pneumonia.6

More recently, H. alvei has featured in only a few case reports in the last decade. Two of these showed similarities to the cases described in our series and some report alternate presentations. First, two papers reported extrapulmonary infections; one a deep soft tissue infection following an open fracture in a 12-year-old boy, and the other a urinary tract infection (UTI) in an 75-year-old man in 2019.21 22 In the latter case, the patient presented with a dry cough, yet H. alvei was only isolated from the urine. Gunthard and Pennekamp’s paper noted only one H. alvei infection of the urinary tract in a patient undergoing aortic valve replacement surgery.19 The authors of the 2019 paper interestingly noted that their patient with UTI had recently been treated for endocarditis and was due for an aortic valve replacement.

In the papers featuring pulmonary manifestation, the first report noted a 95-year-old patient with community-acquired pneumonia leading to empyema, with subsequent aspirate culturing H. alvei.23 A second report featured a 58-year-old man with COPD and H. alvei pneumonia.24 Almost all published cases of H. alvei infections display co-morbidity and immunocompromise or immunosenescence.25 However, one paper published in 2018 reported a healthy 11-month-old infant in Nepal who developed H. alvei bronchopneumonia and bacteraemia.26

H. alvei infections serve as a useful reminder of the fundamentals of bacterial pathogenicity and virulence. Despite the presence of many billions of bacteria in the world, only comparatively few are known to infect humans. This is largely due to two broad factors. First, the bacteria’s innate ability to evade the immune system, infect and cause harm to a host, that is, its pathogenicity, and second, the host’s ability to defend against infection at any given time, thereby determining the organism’s virulence.27

We hope that these cases add to the evidence concerning H. alvei infections and to remind clinicians to remain vigilant for atypical organisms causing infections in the multimorbid and immunosuppressed. These organisms could be resistant to the normal empirical antibiotics used to treat community-acquired pneumonia.

Finally, while our first patient recovered after 5 days, seemingly making a full recovery, the second patient relapsed rapidly on cessation of a 7-day antibiotic course. We question whether the ‘tree-in-bud’ changes seen radiologically are indicative of more severe infection requiring longer treatment with antibiotics. There is sparse literature on H. alvei pneumonia and so this has never been explored.

Patient’s perspective.

Below is a perspective from patient 2 several months after discharge:

Prior to my diagnosis, I had been symptomatic for a number of weeks. During this time, I had been to my general practitioner who suspected my symptoms were due to my recent course of weaning steroids for my Crohn’s disease. I often find that differentiating symptoms from my Crohn’s disease and potentially new health problem can often be difficult, given overlaps of symptoms and drug side effects.

On receiving the diagnosis of pneumonia, I was initially surprised, given I had not experienced difficulty breathing or had a persistent cough. My treatment was remarkably effective and after 48 hours of intravenous antibiotic, I generally felt quite well. Following this, my symptoms relapsed, was readmitted for further antibiotics, and since then, I have made a slow but strong recovery. However, now, I still do not feel back to normal and do find myself short of breath at times and with an ongoing cough.

Learning points.

  • In immunocompromised patients, due to multimorbidity and chronic disease, or iatrogenic, the differential diagnosis for shortness of breath and cough is varied and wide.

  • While imaging may be helpful for the majority of patients, our two cases highlight the diverse radiological manifestation of a single pathogen. Imaging can aid in excluding other differentials and guide bronchoscopy to specific lung segments for bronchoalveolar lavage; however, the most useful investigation for suspected infection in the immunocompromised is culture.

  • Consider positive growth of typically non-pathogenic bacteria in the immunocompromised as possibly a causative agent, while in the immunocompetent, these are more likely to reflect commensal growth.

Acknowledgments

We would like to thank the two patients discussed within this case series. They were kind and willing enough to allow us to discuss their cases.

Footnotes

Contributors: AB: Identified cases and provided oversight of writing and editing. JHG: Wrote the main text, including introduction for patient 1 and discussion. CM: Wrote patient 2 case. SW: Editor, and provided suggestions and comments for the discussion.

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.

Competing interests: None declared.

Patient consent for publication: Obtained.

Provenance and peer review: Not commissioned; externally peer-reviewed.

References

  • 1.Klapholz A, Lessnau KD, Huang B, et al. Hafnia alvei. respiratory tract isolates in a community hospital over a three-year period and a literature review. Chest 1994;105:1098–100. 10.1378/chest.105.4.1098 [DOI] [PubMed] [Google Scholar]
  • 2.Ewing WH. Identification of Enterobacteriaceae by biochemical reactions : Edwards and Ewing’s identification of enterobacteriaceae. New York. NY: Elsevier Science Publishing Co. Inc, 1986: 47–72. [Google Scholar]
  • 3.Møller V. Distribution of amino acid decarboxylases in ENTEROBACTERIACEAE1. Acta Pathologica Microbiologica Scandinavica 1954;35:259–77. 10.1111/j.1699-0463.1954.tb00869.x [DOI] [PubMed] [Google Scholar]
  • 4.Real F, Fernández A, Acosta F, et al. Septicemia associated with Hafnia alvei in laying hens. Avian Dis 1997;41:741–7. 10.2307/1592170 [DOI] [PubMed] [Google Scholar]
  • 5.Kew KM, Seniukovich A, Cochrane Airways Group . Inhaled steroids and risk of pneumonia for chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2014;6 10.1002/14651858.CD010115.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Dalen JE. Pulmonary embolism: what have we learned since Virchow? natural history, pathophysiology, and diagnosis. Chest 2002;122:1440–56. 10.1378/chest.122.4.1440 [DOI] [PubMed] [Google Scholar]
  • 7.Petersen P, Boysen G, Godtfredsen J, et al. Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation. The Copenhagen AFASAK study. Lancet 1989;1:175–9. 10.1016/S0140-6736(89)91200-2 [DOI] [PubMed] [Google Scholar]
  • 8.Im JG, Itoh H, Shim YS, et al. Pulmonary tuberculosis: CT findings--early active disease and sequential change with antituberculous therapy. Radiology 1993;186:653–60. 10.1148/radiology.186.3.8430169 [DOI] [PubMed] [Google Scholar]
  • 9.Hatipoğlu ON, Osma E, Manisali M, et al. High resolution computed tomographic findings in pulmonary tuberculosis. Thorax 1996;51:397–402. 10.1136/thx.51.4.397 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Miller WT, Panosian JS. Causes and imaging patterns of tree-in-bud opacities. Chest 2013;144:1883–92. 10.1378/chest.13-1270 [DOI] [PubMed] [Google Scholar]
  • 11.Rossi SE, Franquet T, Volpacchio M, et al. Tree-in-bud pattern at thin-section CT of the lungs: radiologic-pathologic overview. Radiographics 2005;25:789–801. 10.1148/rg.253045115 [DOI] [PubMed] [Google Scholar]
  • 12.Kraft SC, Earle RH, Roesler M, et al. Unexplained bronchopulmonary disease with inflammatory bowel disease. Arch Intern Med 1976;136:454–9. 10.1001/archinte.1976.03630040056012 [DOI] [PubMed] [Google Scholar]
  • 13.Gabazza EC, Taguchi O, Yamakami T, et al. Bronchopulmonary disease in ulcerative colitis. Intern Med 1992;31:1155–9. 10.2169/internalmedicine.31.1155 [DOI] [PubMed] [Google Scholar]
  • 14.Swinburn CR, Jackson GJ, Cobden I, et al. Bronchiolitis obliterans organising pneumonia in a patient with ulcerative colitis. Thorax 1988;43:735–6. 10.1136/thx.43.9.735 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Janda JM, Abbott SL. The genus Hafnia: from soup to nuts. Clin Microbiol Rev 2006;19:12–28. 10.1128/CMR.19.1.12-28.2006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Gelev I, Gelev E, Steigerwalt AG, et al. Identification of the bacterium associated with haemorrhagic septicaemia in rainbow trout as Hafnia alvei. Res Microbiol 1990;141:573–6. 10.1016/0923-2508(90)90021-H [DOI] [PubMed] [Google Scholar]
  • 17.Sharma RK, Boro BR, Borah P. Incidence of caprine pneumonia and associated bacterial species. Indian J Anim Sci 1991;61:54–5. [Google Scholar]
  • 18.Washington JA, Birk RJ, Ritts RE. Bacteriologic and epidemiologic characteristics of Enterobacter hafniae and Enterobacter liquefaciens. J Infect Dis 1971;124:379–86. 10.1093/infdis/124.4.379 [DOI] [PubMed] [Google Scholar]
  • 19.Günthard H, Pennekamp A. Clinical significance of extraintestinal Hafnia alvei isolates from 61 patients and review of the literature. Clin Infect Dis 1996;22:1040–5. 10.1093/clinids/22.6.1040 [DOI] [PubMed] [Google Scholar]
  • 20.Stock I, Rahman M, Sherwood KJ, et al. Natural antimicrobial susceptibility patterns and biochemical identification of Escherichia albertii and Hafnia alvei strains. Diagn Microbiol Infect Dis 2005;51:151–63. 10.1016/j.diagmicrobio.2004.10.008 [DOI] [PubMed] [Google Scholar]
  • 21.Litrenta J, Oetgen M. Hafnia alvei: A new pathogen in open fractures. Trauma Case Rep 2017;8:41–5. 10.1016/j.tcr.2017.01.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Yarlagadda K, Shrimanker I, Nookala VK. Catheter-associated Hafnia alvei-induced urosepsis. Cureus 2019;11:e6471. 10.7759/cureus.6471 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Loya MK, Walsh J. A case of community acquired pneumonia caused by Hafnia alvei developing into an empyema. Chest 2015;148:101A 10.1378/chest.2223426 [DOI] [Google Scholar]
  • 24.Karamanlı H, Özer TT. A rare case of Hafnia alvei pneumonia presenting with chronic obstructive pulmonary disease. Turk Gogus Kalp Dama 2017;25:308–11. 10.5606/tgkdc.dergisi.2017.12905 [DOI] [Google Scholar]
  • 25.Gruver AL, Hudson LL, Sempowski GD. Immunosenescence of ageing. J Pathol 2007;211:144–56. 10.1002/path.2104 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Nayak N, Baral N, Shrestha R, et al. Hafnia alvei bacteremia following bronchopneumonia in an eleven month old child: a case report from a tertiary care hospital in Nepal. IJALSR 2018;1:22–5. 10.31632/ijalsr.2018v01i02.004 [DOI] [Google Scholar]
  • 27.Isenberg HD. Pathogenicity and virulence: another view. Clin Microbiol Rev 1988;1:40–53. 10.1128/CMR.1.1.40 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from BMJ Case Reports are provided here courtesy of BMJ Publishing Group

RESOURCES