Skip to main content
Download PDF

Pooled prevalence of lymphopenia in all-cause hospitalisations and association with infection: a systematic review and meta-analysis

BMC Infectious Diseases volume 23, Article number: 848 (2023) Cite this article

Abstract

Background

Lymphopenia is defined as a decrease below normal value (often 1.0 x 109 cells/L) of blood circulating lymphocyte count. In the general population, lymphopenia is associated with an increased risk of hospitalisation secondary to infection, independent of traditional clinical risk factors. In hospital, lymphopenia is associated with increased risk of healthcare-associated infection and mortality. By summarising lymphopenia’s prevalence and impact on clinical outcomes, we can identify an at-risk population and inform future studies of immune dysfunction following severe illness.

Methods

Peer-reviewed search strategy was performed on three databases. Primary objective was to summarise the pooled prevalence of lymphopenia. Primary outcome was infection including pre-existing lymphopenia as a risk factor for admission with infection and as an in-hospital risk factor for healthcare-associated infection. Secondary outcomes were length of stay and mortality. Mortality data extracted included in-hospital, 28/30-day (‘early’), and 90-day/1-year (‘late’) mortality. Meta-analysis was carried out using random-effects models for each outcome measure. Heterogeneity was assessed using I2 statistic. Joanna Briggs Institute checklist for cohort studies was used to assess risk of bias. The protocol was published on PROSPERO.

Results

Fifteen observational studies were included. The pooled prevalence of lymphopenia in all-cause hospitalisations was 38% (CI 0.34-0.42, I2= 97%, p< 0.01). Lymphopenia was not associated with an infection diagnosis at hospital admission and healthcare associated infection (RR 1.03; 95% CI 0.26-3.99, p=0.97, I2 = 55% and RR 1.31; 95% CI 0.78-2.20, p=0.31, I2=97%, respectively), but was associated with septic shock (RR 2.72; 95% CI 1.02-7.21, p=0.04, I2 =98%). Lymphopenia was associated with higher in-hospital mortality and higher ‘early’ mortality rates (RR 2.44; 95% CI 1.71-3.47, p < 0.00001, I2 = 89% and RR 2.05; 95% CI 1.64-2.56, p < 0.00001, I2 = 29%, respectively). Lymphopenia was associated with higher ‘late’ mortality (RR 1.59; 1.33-1.90, p < 0.00001, I2 = 0%).

Conclusions

This meta-analysis demonstrates the high prevalence of lymphopenia across all-cause hospitalisations and associated increased risk of septic shock, early and late mortality. Lymphopenia is a readily available marker that may identify immune dysfunctional patients. Greater understanding of immune trajectories following survival may provide insights into longer-term poor clinical outcomes.

Peer Review reports

Background

Immune dysfunction plays a central role in a wide range of diseases including cancer, atherosclerosis, trauma, and infections [1,2,3,4]. In the context of sepsis, this dysfunction is demonstrated across innate and adaptive immunity, and is characterised by apoptosis of immune cells, dysfunction in cellular function of neutrophils and monocytes, and ‘T cell exhaustion’ [5,6,7,8,9]. The presence of these cellular dysfunctions is associated with poor clinical outcomes, including healthcare-associated infections, increased mortality, and prolonged hospital length of stay [10, 11]. Immune modulation in cancer through immune checkpoint blockade has revolutionised cancer treatment [12]. There is intense focus on investigating novel therapies to modulate the immune dysfunction in sepsis, in the hope of improving clinical outcomes [13].

There is no one test to identify patients with immune dysfunction and assays are highly specialised and not readily available in a hospital setting [14]. Lymphopenia is a window into the state of the immune system, available from routinely collected clinical data. Lymphopenia has been associated with increased mortality and healthcare-associated infection amongst patients with sepsis [15]. Adverse outcomes associated with lymphopenia have been recognised in a wider hospital setting, including in patients with pneumonia and following stroke [16, 17]. In the general population, lymphopenia is associated with an increased risk of hospitalisation secondary to infection, independent of clinical risk factors such as age and co-morbidities [18, 19].

The ability to identify and quantify this at-risk population is important for designing future studies to modulate the immune response and to investigate the longer-term impact of immune dysfunction in hospital. Since lymphopenia has been shown to lead to poor clinical outcomes in such a wide range of hospital populations, we sought to summarise the pooled prevalence of lymphopenia in hospitalised patients regardless of the cause of hospital admission. In addition, we aimed to determine the impact of lymphopenia on clinical outcomes including infection, mortality and length of hospital stay.

Methods

Protocol and registration

The systematic review was registered prospectively with PROSPERO (CRD42022327031) and was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines [20, 21].

Study search strategy

PROSPERO and Cochrane Library searches confirmed that there were no previous systematic reviews of prevalence of lymphopenia in all-cause hospital admissions. Searches were performed on MEDLINE, Embase and CENTRAL databases. Studies that allowed extraction of prevalence data of lymphopenia were included. Records were not restricted by publication date. Records were extracted to Endnote (Thompson, Reuters, Philadelphia, PA, USA) to remove duplicate studies.

Rayyan was used for title and abstract screening [22]. A sample of 10% of results were reviewed by two reviewers to establish agreement (ZCE and TPH). Any disagreements were re-evaluated and resolved between the two reviewers. Data extraction was carried out by a single reviewer (ZCE).

Exclusion criteria

Narrative reviews, editorials, case reports, duplicate publications, qualitative studies, conference abstracts, and non-human studies were excluded. Studies were limited to adult populations and those published in English language. A protocol amendment, prior to formal screening of search results/data extraction, was published on PROSPERO to exclude studies where the primary focus were patients with pre-existing immunosuppression, HIV or COVID-19. This review aimed to summarise lymphopenia in the general hospital population, rather than in patients with immunosuppression (for example, secondary to chemotherapy), in whom lymphopenia and subsequent infection risk are well recognised. Studies of COVID-19 patients were excluded because lymphopenia in this context has been summarised in a recent systematic review [23]. This protocol update did not require a secondary amendment of the search strategy, as negative searching was not carried out.

Data collection process

Data were extracted from the selected papers onto a pre-formatted Excel worksheet (Microsoft, Redmond, WA, USA) containing the following characteristics: author and year of publication; country of origin and setting; study design; duration of study; demographics including age and sex; sequential organ failure assessment (SOFA) score; acute physiology and chronic health evaluation II (APACHE II) score; co-morbidities where available; and the definition of lymphopenia. Outcomes reported were extracted including healthcare-associated infection, all-cause mortality, and length of stay. The diagnostic criteria used for infection and healthcare-associated infection were also extracted.

The number of events was extracted for dichotomous outcomes. Means with standard deviation (SD) were extracted for continuous outcomes. Median values, interquartile ranges and sample size were used to estimate the sample mean and SD [24, 25].

Outcomes

The primary objective was to summarise the pooled prevalence of lymphopenia in all-cause hospital admissions. The primary clinical outcome was infection, including infection at admission and healthcare-associated infection. Secondary clinical outcomes included length of hospital stay, length of intensive care unit (ICU) stay, all-cause in-hospital mortality, 28/30-day mortality (defined as ‘early’) and 90-day/1-year mortality (defined as ‘late’).

The definition of lymphopenia was determined by the paper being analysed in the review. Absolute lymphocyte count (ALC) is expressed in units of 109/L. The normal range of ALC is often accepted to be between 1.5 to 4 x 109/L.

Risk of bias and quality of evidence assessment

The risk of bias was assessed by two reviewers (ZCE and TPH), using the Joanna Briggs Institute (JBI) critical appraisal checklist for observational studies [26].

An overall assessment of the evidence quality for outcome measures was reported according to the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) assessment [27]. The software program GRADEpro was used [28].

Statistical analysis

The meta-analysis was conducted using Review Manager and R Meta package [29,30,31]. A p value of less than 0.05 was accepted to be statistically significant. Dichotomous data were analysed using risk ratio (RR) with 95% confidence intervals (CIs). Continuous data were analysed using inverse variance (I-V) method to obtain mean difference (MD) and standard deviation (SD). Random-effects models for pooled analysis was used, independent of heterogeneity. Heterogeneity was assessed using the I2 statistic.

Results

Study selection

A total of 6006 studies were identified. After title and abstract screening, 236 potentially eligible studies underwent full-text review. The study flow diagram based on PRISMA guidelines demonstrates reasons for exclusion (Fig. 1). Following exclusion of 221 studies (Fig. 1), 15 studies were included in the analysis [15,16,17, 32,33,34,35,36,37,38,39,40,41,42,43].

Fig. 1
figure 1

PRISMA flow diagram showing literature search results. Fifteen studies were used for meta-analysis. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analysis [21]

Full size image

Study characteristics

A total of 72305 patients were included in the analysis (Table 1). Studies included a range of clinical conditions and settings, including elective admissions, critically ill patients, community-acquired pneumonia (CAP), ventilator-associated pneumonia (VAP), sepsis, spinal surgery, chest trauma, traumatic brain injury, and influenza A. Four studies were related to patients in ICUs only [34, 35, 40, 43].

Table 1 Characteristics of included papers. R* retrospective cohort study; P** prospective cohort study. p values*, where provided, are for comparisons between lymphopenic vs non-lymphopenic populations. HR, hazard ratio; OR, odds ratio; LOS, length of stay; IQR, interquartile range; RCT, randomised controlled trial; COPD, chronic obstructive pulmonary disease; UTI, urinary tract infection; CAP, community-acquired pneumonia; RT-PCR, reverse transcription polymerase chain reaction
Full size table

Nine studies were retrospective observational cohort studies and six were prospective cohort studies (Table 1). Lymphopenia was variably defined in all included papers, with a range of 0.5 to 1.2 x 109/L. Two studies did not report when ALC measures were collected [16, 42]. Eight studies reported ALC measures either on admission or within 24 hours of admission [17, 33, 34, 37,38,39, 41, 43]. Two studies measured ALC on specific days determined a priori; 4th day after a blood culture was taken in septic patients and postoperative day 7, respectively [15, 36]. Three studies measured ALC at multiple time points during hospital admission [32, 35, 40].

Risk of bias

Risk of bias was high amongst most included studies (Fig. 2). Areas of high risk of bias or uncertainty related to three main issues. Firstly, there were significant baseline differences between lymphopenic and non-lymphopenic groups in measures of demographics and disease severity. Secondly, completeness of data or strategies to deal with missing data were not reported. Thirdly, it was often unclear whether participants were free of the outcome at start of the study as these were not always measured in a reliable way.

Fig. 2
figure 2

Risk of bias using the JBI Critical Appraisal Checklist for Cohort Studies [26]. Plot created using robvis software [44]

Full size image

Funnel plots were visually inspected for identification of publication bias, where more than five publications reported a specific outcome. Visual inspection of funnel plots demonstrated high risk of publication bias.

Pooled prevalence of lymphopenia

The pooled prevalence of lymphopenia in all-cause hospitalisations was 38% (random effects model proportion 0.38; CI 0.34-0.42, I2 = 97%, p< 0.01) (Fig. 3a).

Fig. 3
figure 3

a Forest plot of pooled prevalence of lymphopenia (of any definition) across 15 studies. CI, confidence interval. b Subgroup analysis of ICU pooled prevalence of lymphopenia of any definition. c Subgroup analysis of septic shock pooled prevalence of lymphopenia of any definition. d Subgrouping of lymphopenia based on absolute lymphocyte count at time of admission (ALC, x 109/L)

Full size image

Given the significant study-related heterogeneity across all included studies, further subgroup analysis was carried out. Studies investigating an ICU population (n = 1020) demonstrated lymphopenia in 34% of admissions (random effects model proportion 0.34; 95% CI 0.26-0.44, I2 = 88%, p< 0.01) (Fig. 3b), although heterogeneity remained high [34, 35, 40, 43]. Subgroup analysis for septic shock populations (n = 1371) demonstrated higher prevalence of 54% with significant heterogeneity (random effects model proportion 0.54; 95% CI 0.30-0.76, I2 = 99%, p< 0.01) (Fig. 3c) [15, 16, 32, 34, 43].

Lymphopenia definition varied across studies, however eight studies defined lymphopenia based on ALC at hospital admission and/or within 24 hours of admission [17, 33, 34, 37,38,39, 41, 43]. Admission lymphopenia had a prevalence of 39% (random effects model proportion 0.39; 95% CI 0.31-0.47, I2 = 97%, p< 0.01) (Fig. 3d). Heterogeneity remained high.

Lymphopenia and infection

Two studies reported infection as the cause of admission [32, 39]. Lymphopenia was not associated with an infection diagnosis (RR 1.03; 95% CI 0.26-3.99, p=0.97, I2 = 55%) (Fig. 4). Subgroup analysis for this outcome was not possible due to the small number of studies. However, heterogeneity is noted between studies by Andreu-Ballester et al. and Rubio-Rivas et al.; lymphopenia definition (< 1 vs < 1.1 x 109/L), timing of lymphocyte measures (any point during hospital admission vs admission), and study population characteristics including age (> 14 vs ≥ 75 years), respectively (Table 1) [32, 39].

Fig. 4
figure 4

Forest plot of infection (as cause of admission) and lymphopenia (of any definition). CI, confidence interval; M-H, Mantel-Haenszel

Full size image

Seven studies reported the outcome of healthcare-associated infection and lymphopenia [15, 17, 33, 34, 36, 37, 42]. Lymphopenia was not associated with healthcare-associated infection (RR 1.31; 95% CI 0.78-2.20, p=0.31, I2 = 97%) (Fig. 5a). Sensitivity analysis was carried out based on lymphopenia definition, stratified as either less than 1.2 or less than 0.8 x 109/L (Fig. 5b). Heterogeneity was reduced in the analysis of ALC less than 1.2 x109/L but increased for studies of ALC greater than 0.8 x 109/L. The outcome remained non-significant.

Fig. 5
figure 5

a Forest plot of nosocomial infection and lymphopenia (of any definition). Funnel plot analysis demonstrates asymmetric shape. b Forest plot of nosocomial infection and lymphopenia, stratified by lymphopenia definition. Top panel ALC < 1.2 > 0.8 x 109L vs bottom panel ALC < 0.8 x 109L. CI, confidence interval; M-H, Mantel-Haenszel

Full size image

Four studies reported the outcome of septic shock and lymphopenia [16, 32, 34, 43]. Lymphopenia was associated with septic shock (RR 2.72; 95% CI 1.02–7.21, p = 0.04, I2 = 98%) (Fig. 6). Heterogeneity was high between the studies.

Fig. 6
figure 6

Forest plot of septic shock and lymphopenia (ALC stratified by cut-off, x 109/L). CI, confidence interval; M-H, Mantel-Haenszel

Full size image

Lymphopenia and mortality

In-hospital mortality was reported by seven studies (Fig. 7) [17, 32, 33, 37, 39, 40, 43]. Lymphopenia was associated with higher in-hospital mortality (RR 2.44; 95% CI 1.71-3.47, p < 0.00001, I2 =89%) (Fig. 7). Excluding Andreu-Ballester et al.’s study, which analysed data from 58260 hospital admissions, the significant heterogeneity is reduced to 41% (RR 2.13, 95% CI 1.72-2.65, p < 0.00001, I2 =41%) (Fig. 7b) [32].

Fig. 7
figure 7

a Forest plot of in-hospital mortality and lymphopenia (ALC stratified by 1.1 x 109/L as cut-off). Funnel plot demonstrates asymmetry. b Exclusion of Andreu-Ballester et al. changes to I2 = 41%. CI, confidence interval; M-H, Mantel-Haenszel

Full size image

‘Early’ (28/30-day) mortality was reported in six studies [15, 16, 34, 35, 38, 41]. Bermejo-Martin et al. studied two cohorts; ‘derivation’ multisite and ‘validation’ single-site cohorts annotated as [1] and [2], respectively (Fig. 8a) [16]. Lymphopenia was associated with higher early mortality (RR 2.05; 95% CI 1.64-2.56, p < 0.00001, I2 = 29%) (Fig. 8a). ‘Late’ (90-day/1-year) mortality was reported in two studies [15, 34]. Lymphopenia was associated with higher late mortality (RR 1.59; 1.33-1.90, p < 0.00001, I2 = 0%) (Fig. 8b).

Fig. 8
figure 8

a Forest plot of 28/30-day mortality (early) with lymphopenia of any definition. Bermejo-Martin et al. (1) and (2): data from derivation cohort and verification cohort, respectively. Funnel plot demonstrates asymmetry. b Forest plot of 90-day/1-year mortality (late) with lymphopenia of any definition. CI, confidence interval; M-H, Mantel-Haenszel

Full size image

Lymphopenia and length of stay

Hospital length of stay (LOS) was reported in five studies (Fig. 9) [17, 33, 37, 38, 41]. Mendez et al. reported hospital LOS of a population of patients with CAP based on two lymphopenia definitions of ALC ≤ 1 x109/L and ALC ≤ 0.724 x 109/L [38]. LOS has been reported separately in the analysis for these two lymphopenia thresholds (Fig. 9) [38]. The overall mean difference in hospital LOS is 1.25 days (95% CI 0.32-2.18, p = 0.008, I2 = 89%) longer for patients with lymphopenia (Fig. 7).

Fig. 9
figure 9

Forest plot of hospital Length of Stay (LOS) and lymphopenia. Mendez et al. (1) and (2): absolute lymphocyte count cut off less than 1.0 x109/L vs. less than 0.724 x109/L, respectively. a Lymphopenia defined ALC less than 1 x 109/L. b Lymphopenia defined as ALC < 0.724 x109/L. CI, confidence interval; I-V, Inverse Variance method

Full size image

Sensitivity analysis was performed based on lymphopenia definition. Four studies, defining lymphopenia as less than 1 x 109/L, demonstrated that lymphopenic populations stayed in hospital 1.85 days (95% CI 1.03-2.66, p < 0.0001, I2 = 74%) longer than non-lymphopenic populations [17, 33, 37, 38]. Two studies, defining lymphopenia as less than 0.724 x 109/L, demonstrated that lymphopenic patients stayed in hospital for 0.35 days longer than the non-lymphopenic population, however this was not statistically significant (95% CI -0.10-0.81, p = 0.13, I2 = 0%) [38, 41].

ICU LOS was reported by three studies (Fig. 10) [15, 33, 37]. Mean difference was 0.17 days longer for non-lymphopenic subgroup, however this was not statistically significant (95% CI -0.34-0.68, p= 0.50, I2 = 60%) (Fig. 10).

Fig. 10
figure 10

Forest plot of ICU LOS and lymphopenia (of any definition). CI, confidence interval; I-V, Inverse Variance method

Full size image

Discussion

This systematic review adds to the growing evidence that lymphopenia is associated with adverse clinical outcomes. We demonstrate that lymphopenia is common in hospitalised patients, occurring in 38% of patients. We demonstrate that lymphopenia is associated with increased early and late mortality. In addition, there is prolonged hospital stay. The analysis did not demonstrate a significant difference in risk of admission with an infection or acquiring a hospital-acquired infection if lymphopenic. However, there was an increased risk of septic shock in lymphopenic patients. The fifteen studies included demonstrate that lymphopenia is seen across a wide range of pathologies including infection, trauma, and intracranial haemorrhagic conditions.

Lymphopenia has been associated with increased mortality and infection risk in a wide range of settings including community populations, perioperative, and sepsis [15, 18, 19, 45]. The studies have broadly shown that lymphopenia is associated with an increased risk of infections and mortality. Given the range of clinical settings in which lymphopenia has been shown to result in adverse clinical outcomes, we summarised for the first time, the prevalence of lymphopenia in all-cause hospitalised patients. Our findings are broadly in line with other studies. In a meta-analysis of peri-operative patients, lymphopenia was associated with a three-fold increase in mortality and a higher rate of postoperative complications and infections [45]. While we demonstrated increased mortality in both ‘early’ and ‘late’ deaths, we did not demonstrate an increase in risk of infection. An increased risk of infection seems intuitively associated with lymphopenia. Lymphopenia is a hallmark of immune dysfunction in sepsis and is associated with healthcare-associated infections [46]. In a single centre observational study, it was persistent lymphopenia lasting beyond the fourth day of sepsis admission, that was associated with a significant increase in secondary infections [15]. Furthermore, in a large population study of 98, 344 individuals, lymphopenia was associated with an increased risk of acquiring infections, including sepsis [18]. When summarising the risk of infection across a broad range of conditions, we did not find a significant association between an infectious cause of hospital admission or healthcare-associated infection. We did, however, show a 3-fold increased risk of septic shock with lymphopenia (RR 2.72; 95% CI 1.02–7.21, p = 0.04, I2 = 98%).

Our review suggests there is a ‘dose-response’ between severity of lymphopenia and adverse clinical outcome. In a retrospective study, Bermejo-Martin et al., identified a subgroup of patients with CAP who were lymphopenic (ALC less than 0.724 x 109/L) that accounted for a significant portion of individuals who developed septic shock and demonstrated a significantly higher risk of 30-day mortality [16]. Consistent with this finding, in a large cohort study, Andreu-Ballester et al. demonstrated that the lowest absolute values were demonstrated in sepsis and septic shock, with severe low absolute counts of lymphocytes associated with higher risk of mortality [32]. Drewry et al. stratified lymphopenia definitions as moderate and severe persistent lymphopenia. This stratification demonstrated a higher incidence of nosocomial infections alongside higher 28-day and 1-year mortality rates in the severe cohort compared to the moderate cohort [15]. These findings suggest a relationship between severity of lymphopenia and outcome, specifically in subgroups of septic shock.

Given the spectrum of conditions that lymphopenia is present in, there is uncertainty whether lymphopenia is an epiphenomenon of an unwell patient or whether it plays a central role in morbidity and mortality. The significance of lymphopenia in different clinical settings and populations is uncertain. Studies to date indicate that lymphopenia reflects a wider dysfunctional immune system. This is certainly shown in studies in sepsis, where immune dysfunction is characterised not only by lymphopenia, but also low monocyte HLA-DR, increased PD-1 and increased regulatory T cells [5,6,7,8,9]. Other routinely measured biomarkers reflect immune dysfunction and have been shown to be associated with increased mortality. In a large population of 31,178 outpatients, in addition to lymphopenia, high levels of C-reactive protein (CRP) were also associated with reduced survival [19]. A follow-up study of sepsis survivors identified a hyperinflammation/immunosuppression phenotype with a significantly higher 1-year mortality risk, demonstrated by CRP as a marker of ongoing inflammation and additional markers of immunosuppression including soluble PD-L1 [10]. Although our review cannot conclude that lymphopenia in the included studies is due to immune dysfunction, our findings are consistent with current understanding of immune perturbations in acute illness.

There are limitations to this study. We aimed to determine the prevalence of lymphopenia in a ‘general’ hospital population. For this reason, we excluded studies that specifically focused on immunosuppressed populations in which the prevalence of lymphopenia and the associated infection risk would be much higher. These populations are immunosuppressed secondary to medical treatments for cancer or inflammatory diseases, and so represent a different population to those with immune dysfunction because of an acute disease. However, it is possible that some patients within the included studies of ‘general’ populations would be on immunosuppressive medications and contribute to the lymphopenic population. In addition, it can be argued that patients with COVID-19 should be represented in the general in-patient population. We excluded these studies because lymphopenia is a well-recognised characteristic and systematic review of lymphopenia in COVID-19 patients has been recently published [23].

This review is further limited by the range of lymphopenia definitions used in the studies, resulting in high levels of heterogeneity in the meta-analysis. Definition of lymphopenia ranged from 0.5 to 1.2 x 109/L. The lack of a unified definition of lymphopenia demonstrates the need for further research in causality, and in clarifying whether there is a potential count-dependent relationship between severity of lymphopenia and outcome.

Lastly, the conclusions made by this meta-analysis are limited by the quality of studies included. Most of the studies had a high risk of bias or uncertainty regarding risk of bias. Since the included studies were observational studies, the GRADE quality of evidence was often downgraded (Fig. 11). However, large sample sizes in studies such as Andreu-Ballester et al., Bermejo-Martin et al., and Campbell et al. allowed upgrading of quality due to large effect size demonstrated across multiple outcomes (Fig. 11) [16, 32, 33].

Fig. 11
figure 11

Summary of findings table and GRADE assessment of outcomes. a Wide confidence intervals for overall effect estimate. b Lack of confounding variables identification [17, 33, 36]. Follow-up time not reported/unclear [15, 36]. Significant baseline differences between lymphopenic and non-lymphopenic subpopulations [37]. c Statistically significant high heterogeneity. d Asymmetrical funnel plot. e Ceccato et al. determined lymphopenia cut-off based on previous analysis [34]. f Lack of generalisability to wider hospitalised population; Ceccato et al.'s study population was related to intensive care-related nosocomial infection while Zhou et al.'s study concentrated on patients with severe influenza A patients [34, 42]. g Significant differences between lymphopenic and non-lymphopenic subpopulations [37, 43]. Unclear reporting of characteristics across groups in Andreu-Ballester et al. and Vulliamy et al. [32, 40] h Significant differences between lymphopenic and non-lymphopenic subpopulations [16, 38, 41]. Potential confounding factors not identified [34]

Full size image

In conclusion, this meta-analysis shows that lymphopenia is common across all-cause hospitalisations and associated with increased risk of mortality and length of stay. Moreover, given the consistent findings across several types of pathology, the data suggest a link between lymphopenia at any point during a hospital stay and poor outcome. This meta-analysis highlights the paucity of available high-quality evidence. By summarising the prevalence of lymphopenia in hospitalised patients, this review may inform the design of future studies investigating outcomes and novel treatments for immune dysfunction in hospitalised patients. In particular, prospective studies of lymphocyte count and its potential correlation with detailed immunophenotyping and longer term patient outcomes may provide further insight into the value of lymphopenia as a marker of immune dysfunction and prediction of illness trajectory after hospitalisation.

Availability of data and materials

The datasets generated and analysed during the current study are available below.

Embase search strategy:

https://ovidsp.ovid.com/ovidweb.cgi?T=JS&NEWS=N&PAGE=main&SHAREDSEARCHID=68ESh1P8ZoHE1kmkNz5h48BVP89Yt1WPJaCI8ngQgah8cAWJcroVMyVvC1b3XNmtQ

MEDLINE search strategy:

https://ovidsp.ovid.com/ovidweb.cgi?T=JS&NEWS=N&PAGE=main&SHAREDSEARCHID=7Cj8rfrUOTD7CIIlkeLwDOPZ5ZiSJ1mGva3uMLQ3fCkAiPYdXDsB5ss6FZ0Glvvs2

CENTRAL search strategy:

"lymphopenia" OR lympho?enia OR lymphocyte*

AND

hospital* OR "critical care" OR "intensive care"

AND

infection*

AND

"randomi*ed controlled" OR "randomi*ed clinical" OR "randomi*ed intervention" OR "controlled clinical" OR observational

Abbreviations

(PRISMA) guidelines:

Preferred Reporting Items for Systematic Reviews and Meta-Analysis

(SOFA) score:

Sequential organ failure assessment

(APACHE II) score:

Acute physiology and chronic health evaluation II

SD:

Standard deviation

ICU:

Intensive care unit

ALC:

Absolute lymphocyte count

(JBI) critical appraisal checklist for observational studies:

Joanna Briggs Institute

(GRADE) assessment:

Grading of Recommendations, Assessment, Development and Evaluations

RR:

Risk ratio

OR:

Odds ratio

CIs:

Confidence intervals

(I-V) method:

Inverse variance

(M-H) method:

Mantel-Haenszel

MD:

Mean difference

CAP:

Community-acquired pneumonia

VAP:

Ventilator-associated pneumonia

COPD:

Chronic obstructive pulmonary disease

LOS:

Length of stay

References

  1. Burnham KL, Davenport EE, Radhakrishnan J, Humburg P, Gordon AC, Hutton P, et al. Shared and distinct aspects of the sepsis transcriptomic response to fecal peritonitis and pneumonia. Am J Respir Crit Care Med. 2017;196(3):328–39.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  2. Xiao W, Mindrinos MN, Seok J, Cuschieri J, Cuenca AG, Gao H, et al. A genomic storm in critically injured humans. J Exp Med. 2011;208(13):2581–90.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  3. Sun Y, Tan J, Miao Y, Zhang Q. The role of PD-L1 in the immune dysfunction that mediates hypoxia-induced multiple organ injury. Cell Commun Signal. 2021;19(1):76.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  4. Javeed N, Gustafson MP, Dutta SK, Lin Y, Bamlet WR, Oberg AL, et al. Immunosuppressive CD14+HLA-DRlo/neg monocytes are elevated in pancreatic cancer and “primed” by tumor-derived exosomes. OncoImmunology. 2017;6(1):e1252013.

    Article  PubMed  Google Scholar 

  5. Conway Morris A, Datta D, Shankar-Hari M, Stephen J, Weir CJ, Rennie J, et al. Cell-surface signatures of immune dysfunction risk-stratify critically ill patients: INFECT study. Intensive Care Med. 2018;44(5):627–35.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  6. Shankar-Hari M, Fear D, Lavender P, Mare T, Beale R, Swanson C, et al. Activation-associated accelerated apoptosis of memory B Cells in critically Ill patients with sepsis. Crit Care Med. 2017;45(5):875–82.

    Article  PubMed  Google Scholar 

  7. Venet F, Chung C-S, Kherouf H, Geeraert A, Malcus C, Poitevin F, et al. Increased circulating regulatory T cells (CD4+CD25+CD127−) contribute to lymphocyte anergy in septic shock patients. Intensive Care Med. 2009;35(4):678–86.

    Article  PubMed  Google Scholar 

  8. Drewry AM, Ablordeppey EA, Murray ET, Beiter ER, Walton AH, Hall MW, et al. Comparison of monocyte human leukocyte antigen-DR expression and stimulated tumor necrosis factor alpha production as outcome predictors in severe sepsis: a prospective observational study. Crit Care. 2016;20(1):334.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Tomino A, Tsuda M, Aoki R, Kajita Y, Hashiba M, Terajima T, et al. Increased PD-1 expression and altered T cell repertoire diversity predict mortality in patients with septic shock: a preliminary study. PLoS One. 2017;12(1):e0169653.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Yende S, Kellum JA, Talisa VB, Peck Palmer OM, Chang CH, Filbin MR, et al. Long-term host immune response trajectories among hospitalized patients with sepsis. JAMA Netw Open. 2019;2(8):e198686.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Conway Morris A, Anderson N, Brittan M, Wilkinson TS, McAuley DF, Antonelli J, et al. Combined dysfunctions of immune cells predict nosocomial infection in critically ill patients. Br J Anaesth. 2013;111(5):778–87.

    Article  PubMed  CAS  Google Scholar 

  12. Rao M, Valentini D, Dodoo E, Zumla A, Maeurer M. Anti-PD-1/PD-L1 therapy for infectious diseases: learning from the cancer paradigm. Int J Infect Dis. 2017;56:221–8.

    Article  PubMed  CAS  Google Scholar 

  13. Francois B, Jeannet R, Daix T, Walton AH, Shotwell MS, Unsinger J, Monneret G, Rimmelé T, Blood T, Morre M, Gregoire A, Mayo GA, Blood J, Durum SK, Sherwood ER, Hotchkiss RS. Interleukin-7 restores lymphocytes in septic shock: the IRIS-7 randomized clinical trial. JCI Insight. 2018;3(5):e98960. https://doi.org/10.1172/jci.insight.98960.

  14. Pfortmueller CA, Meisel C, Fux M, Schefold JC. Assessment of immune organ dysfunction in critical illness: utility of innate immune response markers. Intensive Care Med Exp. 2017;5(1):49.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Drewry A, Samra N, Skrupky L, Fuller B, Comptona S, Hotchkiss R. Persistent Lymphopenia after diagnosis of sepsis predicts mortality. Shock. 2014;42(5):383–91.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Bermejo-Martin J, Cilloniz C, Mendez R, Almansa R, Gabarrus A, Ceccato A, Torres A, Menendez R, NEUMONAC group. Lymphopenic Community Acquired Pneumonia (L-CAP), an Immunological Phenotype Associated with Higher Risk of Mortality. EBioMedicine. 2017;24:231–6.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Carneiro T, Spears W, LeClair J, Mahoney T, Takahashi C, Cervantes-Arslanian A. Admission Lymphocytopenia is Associated with Urinary Tract Infection and Nosocomial Infections in Hemorrhagic Stroke. J Stroke Cerebrovasc Dis. 2021;30(11):106079. https://doi.org/10.1016/j.jstrokecerebrovasdis.2021.106079.

  18. Warny M, Helby J, Nordestgaard BG, Birgens H, Bojesen SE. Lymphopenia and risk of infection and infection-related death in 98,344 individuals from a prospective Danish population-based study. PLoS Med. 2018;15(11):e1002685. https://doi.org/10.1371/journal.pmed.1002685.

  19. Zidar DA, Al-Kindi SG, Liu Y, et al. Association of Lymphopenia With Risk of Mortality Among Adults in the US General Population. JAMA Netw Open. 2019;2(12):e1916526. https://doi.org/10.1001/jamanetworkopen.2019.16526.

  20. Prevalence of lymphopenia in all hospitalisations and its association with infection: a systematic review. PROSPERO 2022 CRD42022327031. 2022. Available from: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42022327031.

  21. Page MJMJ, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, The PRISMA, et al. statement: an updated guideline for reporting systematic reviews. BMJ. 2020;2021:472.

    Google Scholar 

  22. Ouzzani M, Hammady, H., Fedorowicz, Z., and Elmagarmid, A. Rayyan—a web and mobile app for systematic reviews. Syst Rev. 2016;5.

  23. Huang I, Pranata R. Lymphopenia in severe coronavirus disease-2019 (COVID-19): systematic review and meta-analysis. J Intensive Care. 2020;8:36.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Luo D, Wang X, Liu J, Tong T. Optimally estimating the sample mean from the sample size, median, mid-range and/or mid-quartile range. Stat Methods Med Res. 2018;27(6):178–1805.

    Article  Google Scholar 

  25. Wan X, Wang W, Liu J, et al. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med Res Methodol. 2014;14:135. https://doi.org/10.1186/1471-2288-14-135.

  26. Appraisal JC. JBI Critical Appraisal Checklist for Cohort Studies 2020. Available from: https://jbi.global/critical-appraisal-tools.

  27. Schünemann H BJ, Guyatt G, Oxman A, editors. The GRADE Working Group. GRADE handbook for grading quality of evidence and strength of recommendations. Updated October 2013. Available from: guidelinedevelopment.org/handbook.

  28. GRADEpro GDT: GRADEpro Guideline Development Tool [Software]. 2022. Available from: gradepro.org.

  29. Team R. RStudio: Integrated Development for R. In: RStudio, editor.: PBC, Boston, MA; 2020.

  30. Collaboration TC. Review Manager Web (RevMan Web) [Version 5.3]. Available from: revman.cochrane.org.

  31. Balduzzi S, Rücker G, Schwarzer G. How to perform a meta-analysis with R: a practical tutorial. Evid Based Ment Health. 2019;22(4):153–60.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Andreu-Ballester JC, Pons-Castillo A, González-Sánchez A, Llombart-Cussac A, Cano MJ, Cuéllar C. Lymphopenia in hospitalized patients and its relationship with severity of illness and mortality. PLoS ONE. 2021;16(8):e0256205. https://doi.org/10.1371/journal.pone.0256205.

  33. Campbell B, Budreau D, Williams-Perez S, Chakravart S, Galet C, McGonagill P. Admission lymphopenia predicts infectious complications and mortality in traumatic brain injury victims. Shock. 2022;57(2):189–98.

    Article  PubMed  CAS  Google Scholar 

  34. Ceccato A, Panagiotarakou M, Ranzani OT, Martin-Fernandez M, Almansa-Mora R, Gabarrus A, Bueno L, Cilloniz C, Liapikou A, Ferrer M, Bermejo-Martin JF, Torres A. Lymphocytopenia as a Predictor of Mortality in Patients with ICU-Acquired Pneumonia. J Clin Med. 2019;8(6):843. https://doi.org/10.3390/jcm8060843.

  35. Chung K, Chang HT, Lo SC, Chang LY, Lin SY, Cheng A, Huang YT, Chen CC, Lee MR, Chen YJ, Hou HH, Hsu CL, Jerng JS, Ho CC, Huang MT, Yu CJ, Yang PC. Severe lymphopenia is associated with elevated plasma interleukin-15 levels and increased mortality during severe sepsis. Shock. 2015;43(6):569–75.

    Article  PubMed  CAS  Google Scholar 

  36. Imabayashi H, Miyake A, Chiba K. A novel approach for identifying serologivcal markers indicative of surgical-site infection following spine surgery: Postoperative lymphopenia is a risk factor. J Orthop Sci. 2022;27(3):588–93.

    Article  PubMed  Google Scholar 

  37. Koch K, Troester AM, Chevuru PT, Campbell B, Galet C, McGonagill PW. Admission lymphopenia is associated with discharge disposition in blunt chest wall trauma patients. J Surg Res. 2022;270:293–9.

    Article  PubMed  Google Scholar 

  38. Méndez R, Menéndez R, Amara-Elori I, Feced L, Piró A, Ramírez P, Sempere A, Ortega A, Bermejo-Martín JF, Torres A. Lymphopenic community-acquired pneumonia is associated with a dysregulated immune response and increased severity and mortality. J Infect. 2019;78(6):423–31.

    Article  PubMed  Google Scholar 

  39. Rubio-Rivas M, Formiga F, Grillo S, Gili F, Cabrera C, Corbella X. Lymphopenia as prognostic factor for mortality and hospital length of stay for elderly hospitalized patients. Aging Clin Exp Res. 2016;28(4):721–7.

    Article  PubMed  Google Scholar 

  40. Vulliamy P, Perkins ZB, Brohi K, Manson J. Persistent lymphopenia is an independent predictor of mortality in critically ill emergency general surgical patients. Eur J Trauma Emerg Surg. 2016;42(6):755–60.

    Article  PubMed  CAS  Google Scholar 

  41. Wittermans E, van de Garde EM, Voorn GP, Aldenkamp AF, Janssen R, Grutters JC, Bos WJW, Santeon-CAP study group. Neutrophil count, lymphocyte count and neutrophil-to-lymphocyte ratio in relation to response to adjunctive dexamethasone treatment in community-acquired pneumonia. Eur J Internal Med. 2022;96:102–8.

    Article  CAS  Google Scholar 

  42. Zhou F, Li H, Gu L, Liu M, Xue CX, Cao B, Wang C, National Influenza A(H1N1)pdm09 Clinical Investigation Group of China. Risk factors for nosocomial infection among hospitalised severe influenza A(H1N1)pdm09 patients. Respir Med. 2018;134:86–91.

    Article  PubMed  Google Scholar 

  43. Ruiz LA, Serrano L, Pérez S, Castro S, Urrutia A, Uranga A, et al. Impact of severe lymphopenia on the early prediction of clinical outcome in hospitalized patients with pneumococcal community-acquired pneumonia. Infection. 2023;51(5):1319–27. https://doi.org/10.1007/s15010-023-01984-2.

  44. McGuinness LA, Higgins JPT. Risk-of-bias VISualization (robvis): An R package and Shiny web app for visualizing risk-of-bias assessments. Res Synth Methods. 2021;12(1):55–61. https://doi.org/10.1002/jrsm.1411.

  45. Schroth J, Weber V, Jones TF, Del Arroyo AG, Henson SM, Ackland GL. Preoperative lymphopaenia, mortality, and morbidity after elective surgery: systematic review and meta-analysis. Br J Anaesth. 2021;127(1):32–40.

    Article  PubMed  PubMed Central  Google Scholar 

  46. Boomer JS, To K, Chang KC, Takasu O, Osborne DF, Walton AH, et al. Immunosuppression in Patients Who Die of Sepsis and Multiple Organ Failure. JAMA. 2011;306(23):2594–605.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

Download references

Acknowledgements

Many thanks to Dr. Fiona Beyer for methodological support.

Funding

Not applicable.

Author information

Authors and Affiliations

  1. Translational and Clinical Research Institute, Faculty of Medical Sciences, Newcastle University, Newcastle-upon-Tyne, NE2 4HH, UK

    ZC Elçioğlu, W Sendama, J Powell, AJ Simpson, AJ Rostron & TP Hellyer

  2. Faculty of Medical Sciences Library, Newcastle University, Newcastle-upon-Tyne, NE2 4HH, UK

    L Errington & B Metes

  3. Department of Respiratory Medicine, Royal Victoria Infirmary, Newcastle-upon-Tyne, NE1 4LP, UK

    W Sendama & AJ Simpson

  4. Integrated Critical Care Unit, Sunderland Royal Hospital, South Tyneside and Sunderland NHS Foundation Trust, Sunderland, SR4 7TP, UK

    AJ Rostron

  5. Department of Critical Care Medicine, Royal Victoria Infirmary, Newcastle-upon-Tyne, NE1 4LP, UK

    TP Hellyer

Authors
  1. ZC Elçioğlu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. L Errington
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. B Metes
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. W Sendama
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. J Powell
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. AJ Simpson
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. AJ Rostron
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. TP Hellyer
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

The study was conceived by TPH, AR and AJS. Protocol and search strategies were developed by LE, BM, ZCE and TPH. Abstract screening was performed by ZCE and TPH. Data extraction was carried out ZCE. Data analysis was conducted by ZCE and WS. The manuscript was prepared by ZCE, TPH, JP and AR. All authors read and approved final manuscript.

Corresponding author

Correspondence to TP Hellyer.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

Professor AJ Simpson is a National Institute for Health Research (NIHR) Senior Investigator. The views expressed in this article are those of the author(s) and not necessarily those of the NIHR, or the Department of Health and Social Care. The remaining authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Supplementary Table 1.

The pre-specified imputation algorithm for the seven-category ordinal scale.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Elçioğlu, Z., Errington, L., Metes, B. et al. Pooled prevalence of lymphopenia in all-cause hospitalisations and association with infection: a systematic review and meta-analysis. BMC Infect Dis 23, 848 (2023). https://doi.org/10.1186/s12879-023-08845-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12879-023-08845-1

Keywords

BMC Infectious Diseases

ISSN: 1471-2334

Contact us