Although lymphocyte count reduction, routinely defined as less than 1 to 1.5 × 10^9/L, is commonly encountered in clinical monitoring, it is seldom utilized as a diagnostic criterion for ICU admissions. However, early studies have underscored its clinical utility in the urgent diagnosis of bacteremia patients and in predicting postoperative sepsis [31, 50, 51]. Statistics indicate that 74% of septic patients experienced a decline in lymphocyte count (below 1.5 × 10^9/L) within the first 1–2 days, and 56% had not returned to a normal lymphocyte count by days 6–8 [52].

According to a study conducted in China, severe sepsis patients exhibited a lower lymphocyte count (0.8 × 10^9/L, QL=0.50, Qu=1.12) in comparison to the low-risk group (lymphocyte count of 0.9 × 10^9/L, QL=0.65, Qu=1.42) [53]. Another research highlighted significant disparities in the median lymphocyte counts between sepsis patients, critically ill non-sepsis patients, and healthy control groups, registering at 0.9 × 10^9/L, 1.1 × 10^9/L, and 1.8 × 10^9/L respectively, with the interquartile ranges being 0.6–1.3, 0.7–1.6, and 1.4–2.3 × 10^9/L [54]. A large-scale European study corroborated that a lymphocyte count < 1.1 × 10^9/L correlated with a heightened multi-variable adjusted risk ratio of 1.51 (95% CI 1.21–1.89) for sepsis diagnosis, aligning with increased risks of hospital infections and infection-related mortality [55].

In a retrospective study that included 77 sepsis and 23 non-sepsis adult patients, the daily lymphocyte count of the patients was calculated until discharge or death [8]. Lymphocytes exhibited a high diagnosis of sepsis with an area under the curve value of 0.971 (95% CI = 0.916–0.994). The diagnostic efficacy of lymphocytes was more significant than other biomarkers, such as white blood cells, neutrophil count, and procalcitonin. The results also showed that the 28-day mortality rate was 39.66% in patients with persistent lymphocyte counts below 0.76 × 10^9/L, significantly higher than in patients without persistent lymphopenia. ALC is a promising, low-cost, rapid, and readily available biomarker for diagnosing sepsis. High vigilance is required for sepsis when a non-viral infection is suspected and lymphopenia levels are below the optimal threshold (0.76 × 10^9/L) value. In another single-center retrospective cohort study of 335 adult patients with bacteremia and sepsis, complete blood cell counts were recorded for the first four days following the diagnosis of sepsis [6]. On Day 4, the median ALC was significantly higher in survivors than in nonsurvivors (1.1 × 10^3 cells/µl [IQR 0.7, 1.5] vs. 0.7 × 10^3 cells/µl [IQR 0.5, 1.0]). ALC was also found to be independently associated with 28-day survival (adjusted OR 0.68 [95% CI 0.51, 0.91]) and 1-year survival (adjusted OR 0.74 [95% CI 0.59, 0.93]). Thus, persistent lymphopenia on the fourth day after sepsis diagnosis predicts early and late mortality and may serve as a biomarker for sepsis-induced immunosuppression.

It should be noted that some studies demonstrated no significant correlation between persistent lymphopenia and the incidence of hospital-acquired infections in critically ill patients [56]. Conversely, the neutrophil-to-lymphocyte count ratio [57, 58] and the IL-10-to-lymphocyte count ratio [59] offer a more reliable reflection of the severity of sepsis. Additionally, investigations into subpopulations of lymphocytes have furnished valuable insights, wherein specific subsets of T and B lymphocytes displayed variations associated with the gravity of sepsis [60,61,62]. For instance, CD3 + T lymphocyte counts manifested a decreasing trend in patients with early clinical deterioration (0.5 × 10^9/L ± 0.5 versus 0.7 × 10^9/L ± 0.5, p = 0.06) [61]. Although changes in B cells were not as pronounced as those in T cells, there was a significant reduction in CD19 + CD23 + B cells at admission in septic shock patients, followed by a marked augmentation in survivors, implying these cells potentially harbor significant prognostic value [62].

Numerous studies have affirmed a significant correlation between the degree of lymphopenia within 28 days and the incidence of hospital-acquired infections and mortality [6, 9, 38, 40, 41, 63]. There is a link between the failure to normalize lymphocyte levels post-trauma and higher mortality [63]. In parallel, a sustained decline in lymphocyte levels after sepsis diagnosis, mainly due to T-cell reduction, is associated with increased mortality [6]. In septic patients, a notable reduction in γδ T-cell numbers in the intestinal mucosa compared to healthy controls is linked to increased disease severity and higher mortality rates [19]. A recent study revealed that children with severe sepsis and persistent lymphopenia face higher risks of multi-organ dysfunction syndrome or death in pediatric intensive care units, with the lymphopenia persisting as a composite outcome odds ratio of 2.98 (95% CI [1.85–4.02]; p < 0.01) [5]. Specific research further delineates that a severe T-cell reduction can only serve as a prognostic marker for immune suppression when accompanied by a high proportion of circulating immature granulocytes, attributable to the enhanced T-cell apoptosis mediated by MDSC present in immature granulocytes [6, 64, 65].

Lymphopenia is also hypothesized to be a potential cause or consequence of CAP, potentially associated with chronic illnesses, critical illnesses, lymphocyte adherence to vascular endothelium, or extensive migration to the lungs [45]. A lymphocyte count below 1,000/µL may be an independent biomarker for a 30-day mortality rate in CAP patients. For patients in the L-CAP study, a lymphocyte count under 724/mm³ doubled the risk of in-hospital death within 30 days [45]. Therefore, lymphopenia markedly correlates with higher ICU admission risks and in-hospital and 30-day mortality rates in CAP patients compounded with sepsis. Early identification of lymphopenia may facilitate pinpointing CAP septic patients who are in urgent or impending need of intensive care [7]. A significant correlation was observed between a lymphocyte count below the usual lower limit and the mortality rate in a study involving 3099 COVID-19 patients. Specifically, the lymphocyte count was exceedingly low (less than 5% of white blood cells) in deceased individuals. In contrast, it remained within normal ranges (more than 20% of white blood cells) in survivors with moderate illness [11].

L-CAP is characterized by CD4 depletion, a higher inflammatory response, and low IgG2 levels that correlate with greater severity at presentation and worse prognosis [66]. Earlier, both reduced T-lymphocyte counts and reduced IgG levels were reported to be associated with the poor prognosis of severe CAP [67]. In addition, the researchers also found that the severity of CAP was associated with decreased serum immunoglobulin levels (including total IgG, IgG1, and IgG2) and that low levels of immunoglobulins were independently associated with ICU admission and 30-day mortality [68]. Lymphopenia leading to reduced immunoglobulin levels is predictable. A recent study found that TNF-α was significantly lower in gradable lymphopenia compared to non-lymphopenia and that a lower TNF-α baseline may lead to a reduced number of B-cells and immature B-cells, which can result in lower immunoglobulin levels [69]. Immunoglobulin deficiency is also an indicator of severity in COVID-19 patients, and i.v. administration of IgG can mitigate virus-induced immunosuppression and provide passive immune protection against a broad range of pathogens [70].

The results of the above studies collectively show that lymphopenia is strongly associated with higher mortality rates, substantiating its role as a predictive biomarker and potential therapeutic target for critically ill patients [71]. Notably, in studies significantly correlated with the probability of death on day 28, over 30% of patients exhibited lymphopenia, correlated with aggravated prognosis associated with profound T-cell lymphopenia [61].

The reduction in T lymphocytes holds predictive value for higher risks of secondary infections and prolonged hospital stays [40, 61, 72, 73]. Lymphopenia is closely associated with secondary infections and complications, including infections in the lungs, urinary tract, bloodstream, and abdomen [6]. Distinct from the intrinsic functional status of individual T cells, the heightened susceptibility to secondary infections due to lymphopenia mainly manifests through extensive adaptive cell apoptosis, reduced T-cell diversity [74], and an immune suppression state induced by the relative increase in suppressive cells. The diversity of T-cell receptors (TCR) is critical to ensuring an individual’s effective immune responses to various foreign antigens in a fluctuating environment. Research indicates that the loss of TCR diversity is induced by a reduction in lymphocytes, which aligns with the study outcomes showing a decrease in the proportion of Naïve T cells rather than a reduction in memory T cells [21, 39, 75].

Consequently, the discussion on the decline in lymphocytes continues to encompass the loss of TCR diversity, particularly from thymic atrophy caused by cellular apoptosis. It has been discovered that there is a notable decline in TCRβ diversity in adult patients with sepsis, correlating with an increased mortality rate [76, 77]. In the context of long-term use of invasive devices such as intubation, catheterization, and central venous catheterization, reduced TCR diversity could potentially heighten susceptibility to hospital infections related to certain chronic viral infections (e.g., hepatitis C virus), which are more prone to immune evasion [39, 78]. In addition, 42.7% of sepsis patients harbor multiple viruses [54], considering the cellular exhaustion, reduced CD4 and CD8 T cells, and increased myeloid-derived suppressor cells and regulatory T cells, all of which could facilitate viral reactivation due to the inability to eliminate high pathogen loads thoroughly. The association between lymphopenia and the progression and outcome of sepsis is summarized in Table 2.

Immune system homeostasis depends on the balance between immune cell proliferation and death. In sepsis-associated immune disorders, a persistent decrease in absolute lymphocyte counts is essential for assessing immunosuppression and poor prognosis [79]. The lymphocyte population undergoes apoptosis upon the onset of sepsis, significantly reducing the number of B cells and CD4 + and CD8 + T cells [71]. Meanwhile, peripheral blood lymphocyte counts in ICU survivors began to rebound within 72 h after admission to the ICU [32]. Lymphocyte counts in surviving septic mice returned to normal levels at approximately three weeks post-infection [42]. Thus, the homeostasis of lymphocyte counts in the septic host undergoes a biphasic process of disruption and repair. Lymphocytes undergo a predominantly apoptotic “loss”, while multiple factors regulate their expansion capacity. Patients with sepsis whose lymphocyte counts are not effectively restored may experience excessive apoptosis and proliferative dysfunction of lymphocytes (Fig. 2).

Increased apoptosis and impaired proliferation of lymphocytes contribute to sepsis-induced lymphopenia. TGF-β and catecholamine release or L-arginine depletion during sepsis enhance lymphocyte apoptosis. Reduced thymic output, bone marrow “void”, or decreased levels of thyroid hormones limit the effective proliferation of lymphocytes after the onset of septic lymphopenia. Immune checkpoints and inhibitory cells, including regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs), induce apoptosis and impair proliferation

Apoptosis is pivotal in reducing lymphocytes, as corroborated by several comprehensive studies [76, 80]. The T lymphocytes’ abnormal activation (either hyperactivation or suppression) is closely associated with alterations in subpopulation differentiation, predominantly influenced by their apoptotic regulatory mechanisms [19, 81]. The Fas/FasL pathway, the mitochondrial pathway, and the endoplasmic reticulum stress (ERS)-mediated apoptosis are both involved in extensive lymphocyte apoptosis (Fig. 3a) [82, 83]. The activation of Caspase family proteases is involved in the terminal execution field of lymphocyte apoptosis [84, 85].

The Fas/FasL signaling pathway mediates lymphocyte apoptosis in organ tissues [86, 87] and significantly contributes to apoptotic damage. Activated T cells highly express Fas receptors, especially cytotoxic T lymphocytes strongly activated by antigen; after killing Fas-expressing target cells, they also kill their Fas-expressing companions. In addition, soluble FasL shed on the cell membrane kills its cells or neighboring activated T cells in an autocrine and paracrine manner, referred to as reactivation-induced cell death (RICD) [88]. RICD is induced by an exogenous signaling transduction pathway through the caspase-8, which triggers the downstream executor caspase-3 [89]. Notably, mouse and human CD4 + T cells show differences in sensitivity to Fas-mediated RICD [90, 91]. In contrast, human CD8 + T cells are more sensitive to TNF receptor 1-mediated RICD [92], and the reasons for this warrant further exploration.

Sepsis-induced apoptosis in thymocytes and splenocytes cannot be blocked entirely in Fas receptor or TNF-p55 or TNF-p75 receptor-deficient mice [18, 93], suggesting the existence of other forms of apoptotic pathways. The mitochondrial enzyme activity of T cells in the peripheral blood of septic patients is significantly reduced compared with that of healthy controls [94]. Caspase-9 is a crucial downstream link in the mitochondrial apoptotic pathway, and nonsurvivors of sepsis show positive expression of cleaved caspase-9 in splenic T cells [18]. Besides, the Bcl-2 protein can block thymic and splenic T cell apoptosis by inhibiting the mitochondrial pathway but not the death receptor pathway [35]. Intracellular regulation encompasses the upregulation of pro-apoptotic genes, including Bim, Bid, and Bak, coupled with the downregulation of anti-apoptotic gene Bcl-2 expression [82].

The unfolded protein response (UPR) is usually activated to counteract lymphocyte stress. However, with the progression of sepsis, when the UPR fails to maintain cellular homeostasis, the ERS response is transformed into a pro-apoptotic response, with up-regulation of the critical nuclear transcription factor, C/EBP homologous protein (CHOP), driving apoptosis in splenocytes [83]. In addition, the expression of stimulator of interferon genes (STING) can also lead to apoptosis by triggering ERS in splenic T-cells [95]. In contrast, the Notch intracellular segment can block STING-mediated apoptosis by competing with cyclic guanosine monophosphate-adenosine monophosphate for cyclic dinucleotide binding sites on STING [96].

Extracellular regulation principally involves inflammatory cytokines that partake in the upstream regulatory processes of lymphocyte apoptosis, wherein TGF-β1 may induce lymphopenia through modulating pro-apoptotic pathways [97]. During the septicemic phase, another pronounced characteristic is the sustained and exacerbated activation of the sympathetic nervous system, chiefly manifested as the excessive release of endogenous catecholamines [98]. Catecholamine substances, including dopamine and dobutamine, can foster lymphocyte apoptosis by impacting β receptors on the cell surface [99].

Cell fate after trauma and sepsis may depend on autophagy and apoptosis crosstalk [100]. In a mouse model of CLP, apoptosis of splenic T cells occurs along with reduced levels of autophagy in CD4 + and CD8 + T cells [101, 102]. In addition, mice with lymphocyte-specific knockout of Atg5 or Atg7 showed significantly increased mortality, immune dysfunction, and T-cell apoptosis after CLP [101, 102], and deletion of Atg5 resulted in enhanced levels of the anti-inflammatory cytokine IL-10 after CLP, suggesting that autophagy deficiency is involved in septic immunosuppression.

Other mechanisms of cell death may also be involved in lymphopenia. Necroptosis occurred in T cells lacking caspase-8 on the TNF receptor signaling pathway, suggesting that necrotic apoptosis can be substituted for apoptosis when some factor leads to the failure of apoptotic mechanisms in T cells [103]. Antigen-specific CD4 + and CD8 + T cells lacking glutathione peroxidase four fail to expand, and the T cells rapidly accumulate membrane lipid peroxides with subsequent ferroptosis [104]. Follicular helper T cells are uniquely sensitive to caspase-dependent cellular pyroptosis due to the response to ATP by the ionotropic ATP-gated receptor P2X7 [105, 106]. T-cell pyroptosis can be induced by caspase recruitment domain-containing protein 8 in the resting state. However, pyroptosis is inhibited in the TCR-activated state, suggesting that activation of first signals blocks initial T-cell pyroptosis [107, 108]. Although direct evidence for the above forms of cell death and T-cell loss after sepsis is lacking, it suggests that there may be a more complex regulatory mechanism behind T-cell death in sepsis.

Following a rapid decrease over 1–2 days, the T cell counts in the majority of survivors initiate recovery between days 4 and 7 [61, 66, 77], with CD4 + T cells being the principal contributors [17, 18, 40, 42, 109,110,111]. A poor prognosis is generally observed in patients unable to reach a significant T cell count re