SI was found in 101 of 353 patients in this study. This is a fairly high rate considering the relationship between splenic infarction and infection, which has been of low interest to physicians. In some patients in this study, the infection may simply be a secondary infection. However, we deliberately included these blood vessel compromised statuses. This is because the purpose of this study was to assume the situation when a clinician encounters a patient with splenic infarction and infection, and to inform about its management. To this end, we investigated various etiologies to determine the overall relationship between infection and SI and found that the etiologies were largely divided into the following groups: thrombotic events due to IE or sepsis; events due to the presence of an intracellular organism; and events due to the synergy of a localized infection and other risks.
Thrombogenic events can be considered the main mechanism underlying infection-induced SI (35.3% in this study). Infection can cause thrombosis through various mechanisms. In the presence of inflammatory conditions, increased cytokine production due to sepsis disrupts the coagulation system [12], activating platelets by the action of pro-inflammatory mediators, such as platelet-activating factors [13], as well as the action of P-selectin, which increases systemic inflammation and leads to platelet adhesion [14]. The functioning of the elements of the anti-coagulation mechanism, including anti-thrombin, the protein C system, and inhibitors of the tissue factor pathway, can be compromised by infection [15]. In infectious endocarditis, the common pathogens are Staphylococcus, Streptococcus, and Enterococcus, as well as other fungal species. There may be differences in the degree to which IE is triggered even within a single species. For example, serotypes 2, 5, and 8 of S. aureus are more dangerous [16-18] in thrombotic events. Decreased expression of fibronectin-binding proteins (FnBPA and FnBPB) produces less IE [19], and loss of expression of GspB in Streptococcus is indicated by reduced toxicity of IE [20, 21]. On the other hand, Escherichia and Klebsiella, which are common causes of bacteraemia, have relatively less IE unlike Staphylococcus and the viridans group of Streptococcus [22]. This is because the species have a low ability to attach to a non-bacterial thrombotic embolism. In this study, Escherichia or Klebsiella were found in bacteraemia without IE. In an animal model, administration of lipopolysaccharide purified from Escherichia, Klebsiella, and Salmonella induced a septic state. Host-derived responses to this cell wall component are important in inducing a pro-thrombotic state. The binding of LPS with monocytes/macrophages, platelets, and endothelial cells activates coagulase factors and pro-inflammatory responses, resulting in a pro-coagulant state [23]. Moreover, these pro-inflammatory molecules reduce the levels of anti-coagulant proteins, which can be related with DIC [24, 25]. The occurrence of antiphospholipid syndrome or disseminated intravascular coagulation in association with severe infection can also be considered a thrombogenic event.
In this study, 27 (26.7%) patients were classified with miscellaneous systemic infections, with intracellular organisms representing the causative pathogens in most of these cases. Malaria and babesiosis have previously been reported to cause SI [26, 27]. Both are parasitic diseases in which the parasites infect red blood cells (RBCs). In these diseases, parasitemia and the destruction of RBCs result in hemostasis and cytokine-induced thrombosis, which serve as the main causes of SI; however, the exact mechanism by which SI occurs in these diseases remains unclear. Orientia tsutsugamushi infection was frequently identified in several of our patients. This organism is known to cause endothelial dysfunction, leading to the triggering of fibrin formation and platelet adhesion and aggregation by endothelial cells [14]. Additionally, endothelial dysfunction contributes to the impairment of the protein C system [14]. Antiphospholipid syndrome is another cause of endothelial dysfunction [28]. In addition to the conditions revealed in this study, Q fever, herpes infection, brucellosis, typhoid, and murine typhus [29] are others that cause SI due to endothelial dysfunction. When physicians encounter patients with SI, they must consider the possibility of infection with different pathogens capable of causing endothelial dysfunction. It noteworthy that most of the patients in this group (miscellaneous systemic infection) also had splenomegaly. In addition to endothelial dysfunction, vessel compression accompanying splenomegaly may be considered one of the causes of splenomegaly; however, no studies have clearly delineated this. Identifying the presence or absence of splenomegaly in SI patients may be helpful in differentiating those with intracellular organism(s) from those with SI. In addition, the fact that the patients in the miscellaneous systemic infection group did not exhibit multi-organ occlusion can help differentiate them from other SI patient groups.
Most patients with localized infections and SI had at least one risk factor. Risk factors, including trauma, surgery, pancreatitis, pancreatic tumor, and portal vein thrombosis, which were seen at a rate of 11.9% in this study, are generally considered to be unrelated to infection, except for pancreatic abscess, splenic abscess, and pylephlebitis (liver abscess) [30, 31]. Hematological malignancy, vasculitis, hypercoagulative status, and atrial fibrillation, however, are believed to lead to SI because of their varied associations with infection. Hematological malignancy is believed to be associated with splenomegaly and/or hyperviscosity under acute leukemic conditions. However, infections due to an immunocompromised status should always be monitored, and attention should be devoted to the increased tendency of thrombosis due to infections. Patients with connective tissue disorders may develop SI as a result of vasculitis, splenomegaly, and antiphospholipid syndrome (22). Infections in patients with vasculitis are generally believed to be secondary to an immunocompromised status; nevertheless, the possibility of infection exacerbating connective tissue diseases should also be taken into consideration [32]. Exacerbation of atherosclerosis due to infection is sometimes ignored in clinical practice. Our findings revealed that atherosclerotic disease was the most common risk factor (22/101 [21.8%]) for SI in patients with infection. Several studies have shown that infection exacerbates atherosclerosis [33]. However, acute infection alone is unlikely to cause a sufficiently rapid aggravation of atherosclerosis to cause SI. Other studies suggest that occlusion may be associated with vasospasm. Arterial spasms can be caused by the increased production of the cytokine interleukin-1 and reduced bioavailability of nitric oxide during acute infection (19). In patients with atrial fibrillation, infections can promote SI by enhancing thrombogenic tendencies. Infection can lead to atrial fibrillation; however, only 2 of our patients developed paroxysmal atrial fibrillation, and most patients with atrial fibrillation had experienced it previously. If a clinician detects splenic infarction in patients with localized infection, it is necessary to be careful about multiple factors.
These conditions must be differentiated using various approaches. First, it is important to differentiate between bacteremia and IE. Efforts should then be made to identify the underlying disease(s) or risk factors. If no other risk factors are identified, infection by intracellular organisms should be ruled out. In this study, we found that the presence of intracellular organisms was associated with splenomegaly without infarction in other organs. These findings may help differentiate the causes of various types of SI; nevertheless, a comprehensive evaluation of the patient’s underlying disease and history is important. Furthermore, it is difficult to diagnose SI with a localized infection alone, and in this case, the physician must check whether the patient with SI has other risk factors.
Our study had a few limitations, the first of which was its retrospective design; however, a prospective study investigating this topic would be difficult to perform due to the nature of SI. To overcome the drawbacks of a retrospective study design, we considered various risk factors for SI and analyzed their potential correlations with the development of SI. Second, specific pathogen(s) could not be identified in some of the investigated cases. Third, the findings of this study are based on the experience at a single tertiary hospital; as such, it may be difficult to generalize these findings elsewhere, particularly because diseases, such as malaria and scrub typhus, exhibit significant variation in regional incidence. Fourth, the small number of SI patients was insufficient to address all etiologies. We additionally encountered 2 cases of SI due to Salmonella typhii and Candida albicans after the set study period. It is believed that many more pathogens cause SI. However, the purpose of this study was to examine the nature of the relationship between infection and SI rather than to analyze the exact etiology of SI itself.