Discussion
Tocilizumab has been associated with both atypical and delayed presentations of infections.5 Atypical presentations of common conditions in patients on tocilizumab include case reports of pneumonia with absence of fever and disproportionately normal or mildly elevated biochemical markers.5 Delayed presentations of severe infection have also been reported in patients on tocilizumab, with case reports of disseminated Staphylococcus aureusbacteraemia associated with epidural abscess, polyarticular septic arthritis, and empyema.4 Other unusual presentations include re-activation of latent tuberculosis presenting as fulminant sepsis with splenic abscess two weeks after initiating tocilizumab.6 An increased risk of fungal co-infections and reports of invasive fungal disease have also been observed in patients after a single dose of tocilizumab for the management of COVID-19.7,8
We present a case of Staphylococcus aureus subcapsular splenic abscess and associated left-sided empyema in the setting of tocilizumab. Using “tocilizumab” and “splenic abscess” search terms on PubMed and Google Scholar, there was a single case report of splenic abscess in context of reactivated tuberculosis.6 There were no previously reported cases of non-mycobacterial splenic abscess associated with tocilizumab therapy, and hence this would be the first reported case to the best of our knowledge.
In this clinical case, we hypothesise the formation of splenic abscess was by haematogenous seeding from transient but uncapturedStaphylococcus aureus bacteraemia in the context of antibiotic therapy suppression, with the recurrent Bartholin cyst infection being the likely primary source of infection. The left sided empyema/pleural effusion with biopsy-proven pleural infection, was likely a continuation of the adjacent splenic abscess Staphylococcus infection through the presence of otherwise clinically-insignificant congenital diaphragmatic defects,9 given the similar radiological appearance (Figure 1). While the rotational injury of the spine which prompted the patient’s presentation may have coincidentally caused concurrent musculoskeletal pain in her left upper quadrant and flank, it is likely to be a red herring in the setting of the splenic abscess being the more plausible cause for progressively worsening pain. The significant immunosuppressive regimen of leflunomide, methotrexate and tocilizumab likely contributed to the development of the atypical infection and delayed presentation, due to the patient having a markedly suppressed ability to mount an immune response and hence remaining relatively asymptomatic during the early stages of infection. However, we postulate that this atypical presentation was in particular due to the recent commencement of tocilizumab – given the patient had previously tolerated methotrexate and leflunomide for two years prior without developing significant infections.
This case emphasises the importance of considering infection as a differential diagnosis for common presentations such as suspected musculoskeletal pain in patients who are on tocilizumab. Additionally, given tocilizumab inhibits IL-6 mediated production of CRP,5 a normal CRP may be unreliable for excluding infection. Conversely, elevation in CRP in the setting of active tocilizumab therapy may be a marker of severe disseminated infection ­– as exemplified by the present case. As the availability and use of tocilizumab increases, it is increasingly important for clinicians to have a high index of suspicion for delayed presentations of unusual infections in patients on tocilizumab therapy, including those who present with otherwise common presentations such as suspected musculoskeletal pain.