The spleen is a secondary lymphoid organ responsible for surveillance against pathogens circulating in the blood (Bohnsack and Brown, 1986). Its microscopic structure consists of two compartments: white pulp (WP) and red pulp (RP), with a marginal zone (MZ) between the WP and RP. The splenic compartments undergo substantial changes during infections (Mebius and Kraal, 2005). The WP, responsible for initiating the immune response, usually presents lymphoid follicle (LF) hyperplasia and germinal center (GC) formation. Among other changes characteristic of different diseases, the memory cell pool may increase in the MZ, while the plasma cell population can rise in the RP (Hermida et al., 2018; Lewis et al., 2019).

In long-lasting infections, such as visceral leishmaniasis (VL), the spleen presents sequential changes that result in a large replacement of cell populations, followed by hyperplasia and/or atrophy, as well as the disruption of splenic compartments (Santana et al., 2008; Silva et al., 2012; Silva-O’Hare et al., 2016; Melo et al., 2021). In acute severe infections, such as bacterial sepsis, changes in the spleen are mostly associated with the elimination of cell populations by different mechanisms of cell death (Hotchkiss et al., 2001; Toti et al., 2004; Ihlow et al., 2021).

While some knowledge has been accumulated on spleen changes in chronic diseases and how this organ may be involved in the maintenance of chronic infections (Hermida et al., 2018; Lenti et al., 2022), little is known about splenic alterations and associated consequences in the context of acute severe infections. An interesting study by Huston et al. (2008) showed that inhibiting cell death in the spleen prevented death in a murine model of experimental sepsis (Huston et al., 2008). This observation was further supported by treatment with cell death inhibitors, such as IL-7 and anti-PD-L1 (Cao et al., 2019). Thus, it follows that understanding how acute severe infections provoke changes in the spleen may aid in designing more effective strategies to better manage patients.

From 2019 to 2022, a disproportionate number of individuals died from COVID-19. The infection, caused by the SARS-CoV-2, evolved with severe acute respiratory syndrome, accompanied by coagulopathy and high immune system activation (Iba et al., 2020; Zanza et al., 2022). In the beginning of this period, approximately 20% of patients with severe COVID-19 died, with a mean time between disease onset and death lasting 17 days (Byrne et al., 2020; Dorjee et al., 2020; Linton et al., 2020). Some autopsy studies indicated that the spleen was severely affected by COVID-19 (Duarte-Neto et al., 2020; Liu et al., 2020). However, no systematic studies have reported on the spleen changes evidenced by SARS-CoV-2 infection, nor speculated about possible implications regarding the course of disease.

Using a minimally invasive autopsy technique, we had the opportunity to collect and study spleen samples from nine patients who died of COVID-19. This study reports profound disorganization of spleen compartments due to severe COVID-19, associated with the decreased density of cell populations. We further endeavored to examine the potential pathways involved in this observed disorganization, including cytokine expression and changes in the extracellular matrix that may have disrupted leukocyte distribution in the organ.

The spleen plays a crucial role in immune response by either mounting an enduring immune response against microorganisms such as encapsulated bacteria, or by diverting the immune response, which favors infection (Mebius and Kraal, 2005). Both roles depend on morphological changes in the spleen compartments and the redistribution of cell populations (Silva et al., 2012; Silva-O’Hare et al., 2016; Hermida et al., 2018). Severe acute or chronic inflammation can provoke the disruption of splenic compartments, which may compromise the patient’s immune response and favor disease progression (Hermida et al., 2018). Here we show that severe COVID-19 led to the intense depletion of leukocyte populations in both RP and WP spleen compartments, in addition to marked reductions in the numbers and T and B lymphocytes; plasma cell counts increased in both compartments, while the number of macrophages was increased only in the WP. Together with cell population alterations, reticular fiber networks appeared fragmented, which further disrupts spleen structure and inhibits normal cell population homing. The densities of TNF- and IL-6-producing cells were increased in both splenic WP and RP.

The lymphocyte depletion observed in COVID-19 mirrors that seen in sepsis. Both conditions are characterized by a highly systemic inflammatory status with disproportionate leukocyte activation and death. Many mechanisms of cell death are activated in COVID-19. Apoptosis mediated by the FAS-FASL pathway, which is sustained by the continual stimulation of cytokines such as TNF and IL-6, may have been involved in the lymphocyte death seen in the spleens of the COVID-19 patients studied herein (Feng et al., 2020). The density of cells producing these two cytokines was increased, and the TUNEL analysis conducted revealed widespread cell death. Furthermore, elevated levels of these two cytokines have been negatively correlated with circulating lymphocyte counts and increased expression of T cell exhaustion markers in patients with COVID-19 (Diao et al., 2020; Arcanjo et al., 2021; Zhou et al., 2022). In sepsis, changes in the spleen play a crucial role in the unbalanced inflammatory response that results in immune system exhaustion and death (Cao et al., 2019). Furthermore, interventions such as splenectomy, immunomodulatory treatments, and the inhibition of apoptosis in the spleen have been shown to improve survival in experimental models of sepsis (Huston et al., 2008; Chen et al., 2021). It is important to note that all the COVID-19 patients in this study met the criteria for a diagnosis of sepsis. In fact, severe COVID-19 was reported as the leading cause of sepsis-related deaths from 2020 to 2022, surpassing bacterial sepsis (Karakike et al., 2021; Shappell et al., 2023). Of note, three of these patients also had evidence of coinfection by bacteria and fungus. Although consistent with other studies, we cannot rule out that some TUNEL-positive changes reflect the 5–8 h post-mortem interval, which represents a limitation of this study.

Viral infections alone can provoke disorganization in the spleen structure, as shown in patients with HIV infection or experimental models involving Rhesus Macaques or mice (Falk et al., 1988; Diaz et al., 2002; Moukambi et al., 2015; Samal et al., 2018). In the cases examined here, lymphoid compartments were scarcely visible, with indistinct structural organization. No secondary follicles were observed. In chronic infections such as visceral leishmaniasis, lymphoid follicle disorganization results in the impairment of lymphocyte migration into the lymphoid follicle secondary to follicular dendritic cell death, as well as decreased CXCL13 production (Silva et al., 2012). In such cases, evidence exists that lymphocyte repositioning is responsible for WP disorganization. In COVID-19, the absence of secondary lymphoid follicle formation may be due to the shift from traditional follicular helper differentiation to a Th1 phenotype in response to high expression of TNF in the tissue environment (Feng et al., 2020; Kaneko et al., 2020). Furthermore, the marked lymphocyte death seen in these fatal cases of COVID-19 was likely linked to lymphoid compartment atrophy and disorganization. The observed splenic disorganization and resulting disfunction may favor, together with the anergic state produced by extensive lymphoid cell death, secondary bacterial and fungus coinfections, which could have further aggravated the course of disease, ultimately resulting in death.

Despite extensive WP disorganization, pronounced plasmacytosis with the presence of some Mott cells was noted in the splenic RP. In chronic severe infections such as visceral leishmaniasis, WP disruption and late spleen plasmacytosis have been associated with extended plasma cell survival and splenic homing mediated by increased expression of BAFF, APRIL and CXCL12 (Silva-O’Hare et al., 2016). It follows that changes in stromal and leukocyte cell populations in the spleen due to chronic inflammation may provoke alterations in gene expression patterns. It is possible that altered gene expression may not have occurred as a result of the short (14–26 days) disease duration in our patients with COVID-19. Therefore, the splenic plasmacytosis observed in these patients may be a remnant of a massive follicular antibody response or of continuous extrafollicular plasma cell differentiation. In fact, antibody responses against SARS-CoV-2 infection can occur as early as one-week post-symptom onset, and IgM, IgG, and IgA antibody production has been documented at early stages of infection, possibly via the extrafollicular pathway (Woodruff et al., 2020; Lee et al., 2022). The lack of GC responses in COVID-19, due to increased TNF and IL-6 expression and the rapid progress of WP disruption, result in loss of hypermutation process, which may impair the production of high-affinity antibodies in patients with severe disease (Kaneko et al., 2020; Trevelin et al., 2022). Furthermore, increased levels of TNF and IL-6, key regulators of plasma cell survival, may contribute to the prolonged plasma cell persistence in the spleen (Khodadadi et al., 2019).

Macrophage density was increased in WP but not in the RP of patients with COVID-19. We cannot exclude the possibility that the increased density of these cells in the WP resulted from the extensively decreased lymphocyte population and derangement of the reticular framework of the atrophic lymphoid follicles. Nevertheless, the macrophages present in both spleen compartments appear to be less affected than the other cell types studied. Evidence suggests that the highly inflammatory environment in the spleen may be driven by infected macrophages and dendritic cells, and that these cells play a crucial role in the cytokine storm phenomenon, resulting in lymphopenia (Feng et al., 2020; Merad and Martin, 2020; Abdullaev et al., 2021; Xiang et al., 2021). Conversely, a hyper-responsive immune system in COVID-19 may lead to the excessive activation and proliferation of macrophages, known as macrophage activation syndrome (MAS) (McGonagle et al., 2021). This condition is associated with high levels of ferritin in the blood, blood clotting disorders, splenomegaly and hemophagocytosis (Chen et al., 2023). Three of our patients had extensive iron deposits in the spleen, and all three presented with microcirculatory thrombosis and erythrophagocytosis was evidenced in at least one patient.

Another aspect of spleen disorganization observed in COVID-19 is extensive reticular fiber fragmentation. The spleen is among the organs possessing high regenerative capability. During the regenerative process, proper cell positioning is dependent upon the extracellular matrix. Therefore, the disruption of the splenic framework may constitute an additional impairment to structural regeneration. In severe cases of COVID-19, patients with acute respiratory distress syndrome (ARDS) and low oxygen levels present high levels of metalloproteinases, which can break down the supportive tissue in the lungs, leading to serious organ damage (Zhu et al., 2022; Ma et al., 2023). Similarly, in nervous tissue, elevated cytokine levels, including IL-6—which correlates with cytokine levels observed in the spleen—have been associated with increased protease activity and tissue injury (Fekete et al., 2025). Similar changes may occur in the extracellular matrix of the spleen (Guizani et al., 2021; Lee and Kim, 2022). Alternatively, the cells responsible for producing the fibers in this framework may be directly affected by infection or changes in the splenic environment (Mueller and Germain, 2009; Morgado et al., 2020). Notably, a similar pattern of reticular fiber disruption has also been observed in the spleens of patients with sepsis, likely due to high metalloproteinase activity (Gunia et al., 2006).

In conclusion, the data presented here suggest that the spleen plays a significant role in the inflammatory state seen in COVID-19, whether through direct infection of the organ by SARS-CoV-2 or via systemic inflammation. The observed changes in the spleen are consistent with the patients’ impaired capability to combat the primary infection by SARS-CoV-2, as well as coinfections.