DISCUSSION
To place our case in context, we searched the MEDLINE database on 27 September 2022 to determine the spectrum of pathologies caused byPantoea and associated risk factors. We used the search term ‘Pantoea ’ with no restriction on publication date, and applying the ‘human’ and ‘English language’ filters. This search yielded 278 results, from which 69 case reports, case series, and outbreak reports were selected for data extraction. Patients included in these publications were stratified by immune status and age, with children defined as those <18 years old. We also extracted data on clinical syndromes of sporadic cases, as well as sources of outbreaks. The literature review is summarized in the flowchart shown in Figure 5.
As is evident from the flowchart, Pantoea infections are rare, with only 248 cases reported worldwide between 1991 and 2022, 53 of which were outbreak cases stemming from various contaminated environmental sources.(7-11) Sporadic adult and paediatric cases were seen in equal proportions, with immunocompetent patients greatly outnumbering their immunocompromised counterparts. Bacteraemia,(14-36) followed by skin and soft tissue infections (SSTI),(31, 32, 37-42) were the commonest manifestations of Pantoea infection, but the range of syndromes was wide, including peritonitis,(32, 43-52) bone and joint infection,(32, 53-60) intra-abdominal abscess,(61-63) pneumonia,(31, 64) urinary tract infection,(31, 32) ocular infection,(65-71) and rhinosinusitis.(72) One case each of endocarditis,(15) prosthetic joint infection,(55) and post-neurosurgical meningitis (73) was also reported, as were four cases of neonatal sepsis.(74, 75) Only in 19 patients, all of whom had either SSTI,(37, 41, 42) septic arthritis,(53, 56, 58, 59) or ocular infection (66, 69) was there a history of penetrating plant trauma, indicating that this risk factor is less important than previously assumed. The vast majority (84%) of infections were caused by P. agglomerans , with only four other culprit species reported:P. dispersa ,(16, 17, 25, 27, 28, 72) P. ananatis ,(19, 21, 68) P. calida ,(24, 73) and P. stewartii .(18) In four other cases, identification to species level was unable to be performed,(39, 66, 67, 71) like in ours. However, the diversity of Pantoeaspecies is not adequately differentiated by many laboratory identification methods and, as such, many cases attributed to P. agglomerans may, in fact, have been cause by other species.(76-78)
Our patient, therefore, exhibited a very atypical manifestation of aPantoea SSTI mimicking malignancy, to the extent that both clinical and radiological findings raised concerns for sarcoma. Sarcoma is highly unlikely in our case, given the repeated pure growth ofPantoea spp., the long intervening period without death or deterioration, and the expert opinion provided by the specialist sarcoma multi-disciplinary meeting. While he denied any penetrating trauma, whether plant-related or otherwise, it is possible that micro-abrasions may have occurred when he fell, providing a portal of entry. Furthermore, given his rural residence, it is likely that plant or animal material was present on the ground, and his poor diabetic control likely contributed to the development and progression of infection.
Only one other case of Pantoea pseudotumour has been reported, although this patient from India provided a clear history of penetrating plant trauma due to his work in agriculture.(42) Like our case, this patient also reported a distant history of a fall four years prior to presentation. Regardless, both cases, despite the different management approaches taken, resulted in good outcomes. While surgical drainage of the lesion is ideal, as in the latter case, it is interesting that in our patient the infection was successfully contained with him remaining well ten years after symptom onset, even without surgical management. An important clinical lesson, therefore, is that foreign bodies, especially of plant origin which are not well-visualised on plain radiographs, may be retained following such trauma and act as foci of chronic inflammation leading to pseudotumour formation.
Our case raises some interesting questions. The first is whether the patient’s HTLV-1 infection further predisposes to the establishment of a chronic bacterial infection. HTLV-1, unlike HIV, does not result in overt immunodeficiency, but associations between HTLV-1 and non-bacterial infections, such as scabies and strongylodiasis, are well-described.(79) However, little is known about how HTLV-1 mediates concurrent bacterial infections,(80) making this an important research question of clinical significance for the many infected Indigenous Central Australians in whom rates of bacterial infection far exceed those of their non-Indigenous countrymen,(79) as well as people living with this neglected tropical disease worldwide.
The second question arises from the ability of Pantoea to secrete products with bioremediative and immunogenic potential, facilitating its adaptation to diverse ecological niches, including in hostile environments.(1) It may well be that this has aided the establishment of a well-contained infective focus in our patient, given that macrophage activation and epithelial-mesenchymal transformation due to inflammatory mediators released by P. agglomerans leading to fibrosis has been recently reported.(81) Such a process may have succeeded in walling off the abscess, thus preventing cell-mediated immunity from eradicating the infection but also preventing the development of sepsis. Unfortunately, our laboratory was unable to speciate the causative organism, but other species may also have this capability. As such, research to elucidate the mechanisms of action of pathogen mediators released duringPantoea infections may be clinically useful.