PDF
Citation
Print
INTRODUCTION
Immunosuppression is a major risk factor for the onset of critical illness and subsequent mortality in patients admitted to intensive care units (ICUs) [1]. Generally, immunocompromised individuals have a weakened or impaired immune system that is unable to combat infections and disease progression. Patients admitted to an ICU with acute respiratory failure (ARF) are susceptible to infection owing to invasive treatments, such as mechanical ventilation and central venous catheterization. In such situations, immune-function assessment provides crucial insights into the patient’s vulnerability to infections and other complications and facilitates early intervention.
The immune response in critically ill patients (particularly those with ARF) often rapidly fluctuates, beginning with an excessive inflammatory response that causes tissue damage, followed by immunosuppression, which increases the likelihood of secondary infections and adverse outcomes. These dynamic changes in immune function (known as immunoparalysis) are associated with high mortality rates, highlighting the importance of real-time immune-function monitoring to guide therapy and improve patient outcomes [2]. In ICU settings, reliable assessment of the cellular immune status, particularly through biomarker measurements, can help stratify patients according to their risk of complications and provide more personalized treatment approaches. For example, reduced interferon-gamma (IFN-γ) production (particularly by macrophages, natural killer cells, and T cells) has been correlated with a high risk of infection in immunocompromised individuals [3–5]. Biomarkers reflecting immune status can be used by clinicians to identify patients at high risk and to make informed clinical decisions. However, standardized immune-assessment methods are lacking in ICU settings. Current practices often involve estimating immune function using historical medical data, which may lead to inaccurate assessments of a patient’s current status. This point emphasizes the need for consistent and reliable immune-monitoring tools that can be used to evaluate both innate and adaptive immune pathways, thereby offering comprehensive insights into immune functionality.
The QuantiFERON-Monitor (QFM, Qiagen, Germantown, MD, USA) is a promising immune-function monitoring test for assessing immune responses in transplant recipients [6]. The QFM is used to stimulate both the innate and adaptive immune pathways to provide information on the general immune status of patients. High QFM values indicate robust immune function, whereas low values suggest compromised immunity [7, 8]. However, the potential of the QFM for evaluating immune function in individuals with severe respiratory failure remains insufficiently investigated.
We evaluated the QFM levels of patients with severe ARF upon ICU admission and monitored changes throughout their ICU stay (on day 7 after admission and at discharge). We further examined correlations between QFM results, disease severity, and clinical outcomes to better understand the potential of immune-function assessment for improving patient management in ICU settings.
MATERIALS AND METHODS
Study design and ethical approval
The inclusion criterion of the study was admission to the ICU with ARF. ARF was defined as the inability to maintain adequate gas exchange–characterized by a PaO2 of <60 mm Hg or a PaCO2 of >50 mm Hg–with a pH of <7.35 while breathing room air [9]. The exclusion criterion was that the patient refused to participate in the study. Between July 2020 and October 2021, 171 patients were admitted to the ICU of our hospital, 139 of whom met the ARF criteria. Of them, 40 patients declined to participate, and 99 patients with ARF were included in the final study cohort. This study was conducted prospectively following approval from the Institutional Review Board (IRB) of Pusan National University Yangsan Hospital, Busan, Korea (IRB approval number 05-2020-147). Written informed consent was obtained from all participants. The clinical study was registered at ClinicalTrials.gov (PRE20240510-006) and was conducted in accordance with the Declaration of Helsinki (2013 version).
In addition, we employed criteria for stratifying patients into disease categories, such as acute respiratory distress syndrome (ARDS), coronavirus disease 2019 (COVID-19), pneumonia, and septic shock, following established diagnostic guidelines. ARDS was diagnosed according to the Berlin Definition (2012) [10], which was based on the following criteria: (1) symptoms developed acutely or worsened within 1 week of a known clinical case; (2) observable bilateral pulmonary opacities in chest images (X-ray or computed tomography [CT]), which were not fully caused by effusions, lobar/lung collapse, or nodules; and (3) respiratory failure that was not fully explainable by cardiac failure or fluid overload. COVID-19 was confirmed to detect the presence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in upper respiratory tract specimens using the PowerChek SARS-CoV-2 Real-Time PCR Kit (Kogene Biotech, Seoul, Korea) and the STANDARD M nCoV Real-Time Detection Kit (SD Biosensor Inc., Suwon, Korea). Pneumonia was defined as an acute infection of the lung parenchyma, diagnosed based on the following criteria [11]: (1) the presence of clinical symptoms, such as cough, sputum production, dyspnea, or chest pain, (2) the presence of fever (body temperature >38°C) or hypothermia (body temperature <36°C), (3) elevated or decreased white blood cell counts (>10×109/L or <4×109/L), and (4) radiographic evidence of pulmonary infiltrates on a chest X-ray or CT scan. Septic shock was defined following the 2016 Sepsis-3 criteria, namely: (1) the presence of sepsis, defined as life-threatening organ dysfunction due to a dysregulated host response to infection (indicated by an increase in the sequential organ failure assessment [SOFA] score by ≥ 2 points); (2) persistent hypotension requiring vasopressor therapy to maintain a mean arterial pressure of ≥ 65 mm Hg, despite adequate fluid resuscitation; and (3) a serum lactate level of >2 mmol/L, indicating impaired tissue perfusion [12].
To determine the correlation between disease severity scores and QFM levels, we evaluated and recorded SOFA and Acute Physiology and Chronic Health Evaluation (APACHE) II scores upon admission to our ICU. The SOFA score is used to assess respiratory, cardiovascular, hepatic, coagulation, neurological, and renal systems [13]. Higher scores indicate greater organ dysfunction, and each system is scored on a scale of 0 to 4. The total SOFA score ranges from 0 to 24. We used each patient’s age, chronic health status, and physiological variables (N=12) to determine their APACHE II scores. The total score ranged from 0 to 71, with higher scores indicating more severe disease and a higher risk of mortality [14].
Additionally, the use of respiratory support devices, vasopressors, and renal replacement therapy (RRT) during ICU stay was investigated. Cases requiring these interventions were categorized as severe and showed clear differences in terms of QFM levels. The interventions performed for each condition were based on standard treatment principles and critical care management guidelines, without any specific experimental interventions.
QFM and inflammatory-marker measurements
Blood samples were collected at admission, on day 7, and at discharge during the ICU stay. Samples from 99 patients were collected at the time of ICU admission. Subsequently, samples from 77 patients were collected on day 7, and from 45 patients at the time of ICU discharge.
The samples were transferred to the Department of Laboratory Medicine within 1 hr of collection. Blood samples (1 mL) were placed in appropriate QFM tubes (Qiagen). Within 8 hrs after collection, QFM LyoSpheres (stored at 4°C) containing anti-CD3 (T cell stimulator) and R848 (a Toll-like receptor 7/8 ligand) were added to each QFM tube. The tubes were shaken well until the pellet dissolved completely and were incubated at 37°C for 16–24 hrs. After incubation, the QFM tubes were centrifuged at 2,000–3,000×g for 15 min at room temperature (22±22°C). Subsequently, each QFM tube was stored at 4–27°C for up to 3 days when centrifugation was not immediately possible. After centrifugation, the plasma was either stored at 2–8°C for 28 days or at −70°C for longer durations. Samples stored in either manner were analyzed simultaneously after a sufficient number of patients had been enrolled [15]. Plasma IFN-γ levels (IUs/mL) were measured using the QFM 2-Plate kit ELISA (Qiagen) in an automated DS2 ELISA system (Dynex, Chantilly, VA, USA).
Cytokine measurements
Serum C-reactive protein (CRP) levels were measured using an AU5812 Clinical Chemistry Analyzer (Beckman Coulter, Brea, CA, USA). Serum interleukin-1β (IL-1β) and tumor necrosis factor-α (TNF-α) levels were determined via ELISA, using the Human IL-1β Pre-Coated ELISA Kit (BioGems, Westlake Village, CA, USA) and the Human TNF-α Pre-Coated ELISA Kit (BioGems). These tests were conducted with samples obtained when patients were admitted to the ICU.
Statistical analysis
Continuous variables were expressed as the median and interquartile range (IQR) and analyzed using the Mann–Whitney U-test. The Kruskal–Wallis and Friedman tests were used to assess differences in QFM levels measured at admission, on day 7, and at discharge, among patient groups categorized by disease. Pearson’s correlation coefficient was used to analyze associations between variables, and correlation coefficients were calculated. Univariate logistic regression analysis was performed to identify potential predictors of ICU mortality. Variables with a P of <0.05 in univariate analysis, along with clinically relevant variables, were included in a multivariate logistic regression model to determine independent predictors of ICU mortality. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. Statistical significance was set at P<0.05. Statistical analyses were performed using Statistical Package for Social Sciences software (version 27.0; IBM Corp., Armonk, NY, USA), MedCalc Statistical Software version 18.11.3 (MedCalc, Ostend, Belgium), and R software version 3.6.3 (R Foundation for Statistical Computing, Vienna, Austria).
RESULTS
Clinical characteristics
Table 1 shows the clinical characteristics of 99 patients with ARF enrolled in this study. Among them, 60 (60.6%) were men, and the median age was 65 yrs. Of these patients, 27 (27.3%) were diagnosed with ARDS, 23 (23.3%) with COVID-19, 26 (26.3%) with pneumonia, and 23 (23.2%) with septic shock. Sixty-one patients (61.6%) required mechanical ventilator support, 74 (74.7%) required vasopressors, and 20 (20.2%) received RRT. Eighty-one patients (81.8%) survived their ICU stay, whereas 18 (18.2%) died during their stay.
Five healthy controls, comprising three men (60.0%) with a median age of 70 yrs (IQR 65–75), were also evaluated using the QFM. QFM was the only analytical tool used for the healthy controls; no other analyses were conducted. No symptoms of infection or respiratory failure were observed in the healthy controls.
QFM levels and inflammatory markers at admission
Upon ICU admission, the median QFM levels in patients (5.5 IUs/mL [IQR: 4.0–9.6]) were significantly lower than those in healthy controls (465.0 IUs/mL [IQR: 433.8–516.3]; P<0.001). Overall, 91 patients (91.9%) had a QFM level of ≤60 IUs/mL, among whom 77 (77.8%) had a QFM level of ≤10 IUs/mL. The median CRP, IL-1β, and TNF-α levels of patients at ICU admission were 8.7 mg/dL (IQR: 3.9–14.9), 23.0 (20.0–29.8), and 89.0 pg/mL (IQR: 43.5–118.5), respectively.
When categorized by disease, the QFM levels of patients at ICU admission varied significantly. Patients with COVID-19 or pneumonia showed high median levels of 9.2 IUs/mL (IQR: 5.8–10.3) and 7.9 IUs/mL (IQR: 5.5–54.0), respectively. Patients with ARDS or septic shock showed lower median levels of 4.9 IUs/mL (IQR: 4.0–7.9) and 3.6 IUs/mL (IQR: 3.2–4.1), respectively. No statistical difference was observed between QFM levels in patients with COVID-19 or pneumonia; however, statistical differences were observed among the other categories (Fig. 1). When categorized by disease, the patients’ CRP and IL-1β levels did not exhibit significant differences. However, TNF-α levels differed significantly among patients with ARDS or pneumonia when compared with those in patients with COVID-19 (Fig. 1). No significant association was observed between QFM and inflammatory markers, such as CRP, IL-1β, and TNF-α, at admission (Fig. 2).
QFM levels and disease severities
Severity scores, such as SOFA and APACHE II scores, were not associated with QFM levels at ICU admission (Fig. 3). When categorized in terms of respiratory support device use, patients requiring mechanical ventilation showed lower QFM levels at ICU admission than those receiving high-flow nasal cannula support (P=0.001) or nasal prongs (P<0.001). At ICU admission, the QFM levels did not significantly differ with vasopressor use or RRT.
Changes in QFM levels
QFM levels were measured again on day 7 (N=77) and at ICU discharge (N=45), with median values of 8.3 IUs/mL (IQR: 4.3–21.6) on day 7 and 16.7 IUs/mL (IQR: 4.7–54.0) at discharge. The QFM levels did not significantly differ between the three time points (P=0.230) or between the groups categorized by disease (P=0.053). Fig. 4 shows changes in QFM levels according to disease group for complete cases with results for all three time points (N=43). QFM levels in patients with COVID-19 showed an increasing trend; nevertheless, no statistical significance was found (P=0.310; Fig. 4). On day 7, the QFM levels increased in 46 patients (59.7%) and decreased in 31 patients (40.3%). At discharge, the QFM levels had increased in 16 patients (35.6%), decreased in 28 patients (62.2%), and remained unchanged in one patient (2.2%).
QFM levels at admission and at death in the ICU
QFM levels at ICU admission were significantly lower in patients who ultimately died in the ICU (N=18) than in those who survived (N=81) (4.0 IUs/mL [IQR: 3.2–5.8] vs. 5.8 IUs/mL [IQR: 4.4–10.6], P<0.001). Univariate regression analysis showed that the QFM level at ICU admission was significantly associated with ICU mortality (OR: 0.69, 95% CI: 0.49–0.97, P=0.034). However, this association was not statistically significant in multivariate regression analysis. Changes in QFM levels on day 7, compared with those at ICU admission, whether increased (N=46) or decreased (N=31), were not associated with ICU mortality (Table 2).
DISCUSSION
We demonstrated the utility of using QFM levels as a marker of cellular immune function in patients with severe respiratory failure admitted to the ICU. In our patient cohort, significantly lower QFM levels were observed when compared with those in healthy controls, indicative of compromised immune responses. Furthermore, most patients were discharged from the ICU without achieving normal QFM levels, suggestive of prolonged immune dysfunction. This trend was particularly evident among patients requiring ventilator support, those with septic shock, and those who died in the ICU, all of whom exhibited significantly lower QFM levels. These findings support the use of QFM as a potentially valuable marker for evaluating the immune status of critically ill patients, particularly those with severe respiratory conditions.
The median QFM level observed in this study was 5.5 IUs/mL, which is markedly lower than that reported for healthy individuals (median: 465.0 IUs/mL), aligning with previous results. Mean QFM levels of 555.2 IUs/mL were reported for a study involving healthy participants [6]. In contrast, in another study, a 95th percentile of 463 IUs/mL and a 25th percentile of 82.0 IUs/mL were reported [16]. Additionally, in candidates undergoing lung transplants, the median QFM level was 171 IUs/mL [8], which is considerably higher than that in our ICU cohort. This consistency across studies highlights the impaired cellular immune function observed in our patient cohort, as indicated by their low QFM levels.
QFM levels have shown prognostic implications in patients undergoing transplants and those with COVID-19. For instance, post-transplant patients with QFM levels below specific thresholds (<10 IUs/mL or <60 IUs/mL at 3- or 6-months post-transplantation, respectively) had high risks for opportunistic infections [5, 8]. Similarly, lower QFM levels were associated with severe outcomes in patients with COVID-19. A median IFN-γ level of 4.5 IUs/mL (IQR: 0.8–17.6) was reported for patients with COVID-19 who had pneumonia and respiratory failure [17]. In contrast, infected healthcare workers exhibited much higher levels (median: 537.0 IUs/mL, IQR: 115.5–886.0) than patients in an infectious disease unit (16.3 IUs/mL, IQR: 7.4–50.5) or an ICU (7.1 IUs/mL, IQR: 1.3–48.2) [18]. This contrast highlights the markedly compromised immune response in severe cases of SARS-CoV-2 infection and reinforces the relevance of using QFM levels as a marker of immune impairment. Furthermore, IFN-γ levels below 12.1 IUs/mL have been associated with a high possibility of hospitalization for people with COVID-19 [18]. Consistently, we observed that 91.9% of patients with ARF in the ICU had QFM levels of ≤60 IUs/mL, and 77.8% had levels of ≤10 IUs/mL, further supporting the utility of QFM as an indicator of immune suppression in patients with ARF.
The median QFM levels varied by disease classification: 4.9 IUs/mL for ARDS, 9.2 IUs/mL for COVID-19, 7.9 IUs/mL for pneumonia, and 3.6 IUs/mL for septic shock. Blot et al. [19] also reported that QFM levels did not differ significantly at hospital admission between patients with COVID-19 and those with non-COVID pneumonia, with median levels of 4.4 and 2.6 IUs/mL, respectively. However, Blot et al. [19] found that the production of other inflammatory and anti-inflammatory cytokines, including IL-1β and TNF-α, differed between these groups. In this study, CRP and IL-1β levels did not significantly differ between disease groups. Nonetheless, TNF-α levels differed significantly between the ARDS, pneumonia, and COVID-19 groups. Similarly, Kox et al. [20] reported that serum TNF-α levels were significantly lower in patients with COVID-19 than in those with septic shock and ARDS. TNF-α is a key proinflammatory cytokine produced during infection. Excessive TNF-α production can contribute to cytokine storms, leading to systemic inflammation, tissue damage, and severe complications [21]. These findings further emphasize the heterogeneity of immune responses across different disease states and highlight the importance of examining a broader range of cytokines in conjunction with QFM measurements.
We found that QFM levels did not correlate with traditional inflammatory markers or severity indexes, such as the SOFA or APACHE II scores. However, patients who required ventilator support had significantly lower QFM levels than those requiring less intensive respiratory support, although no significant differences were found based on vasopressor or dialysis requirements. QFM levels may provide unique insights into immune functions that complement traditional severity scores, supporting a potential role in early risk stratification.
Our findings also showed a slight increase in patient QFM levels during ICU stay, ranging from 5.5 IUs/mL at admission to 8.3 IUs/mL on day 7 and 16.7 IUs/mL at discharge. Although QFM levels in patients with COVID-19 increased in this study, the change was not statistically significant. The QFM levels at ICU discharge remained far below normal values. This trend is consistent with previous findings. For instance, Ruetsch et al. [22] reported minimal changes in IFN-γ levels over time among patients with COVID-19. This persistent immune suppression highlights the challenges faced by critically ill patients in achieving immune recovery and suggests that even marginal improvements in QFM levels may not counteract ongoing immune dysfunction.
We found that QFM levels were lower in patients who died in the ICU than in those who survived. Univariate regression analysis revealed that patient QFM levels were associated with ICU mortality; however, the strength of the association was lower after adjusting for other relevant variables, including dialysis and SOFA scores. Consequently, our data indicate that QFM levels varied based on the severity of respiratory failure and correlated with short-term mortality, emphasizing the potential value of QFM as a biomarker for initial severity evaluation [16]. Other reports revealed that the predictive significance of in vitro IFN-γ production in patients with COVID-19 did not differ between survivors and non-survivors, potentially because of the limited number of events [22, 23].
Some study limitations should be considered when interpreting our findings. First, this was a single-center study with a relatively small number of participants, which may have introduced selection bias and limited the generalizability of our findings. Second, the QFM provides valuable insights into immune function; nevertheless, it may not be sufficient as a standalone tool for comprehensive immune assessment. Third, we did not investigate the medications administered to the patients, including glucocorticoids, which can affect IFN-γ secretion. Fourth, bias could have occurred due to a lack of tests performed on day 7 and upon discharge for all patients. Despite these limitations, this study represents the first evaluation of immune function in patients with respiratory failure who were admitted to an ICU based on QFM analysis. Our findings expand upon those reported for previous studies that predominantly focused on patients with COVID-19. Our cohort included individuals with various critical respiratory conditions, such as ARDS, pneumonia, and septic shock, thereby demonstrating the broader applicability of QFM for these patients, where immune dysfunction is often associated with poor outcomes. In future studies, larger, multicenter cohorts should be used to validate the utility of QFM with diverse patient populations and explore its integration with other biomarkers for a more holistic approach to immune assessment.
In conclusion, our findings reveal that patients with severe respiratory failure in an ICU setting exhibited markedly lower QFM levels than healthy individuals, which was indicative of substantial immune dysfunction. Reduced QFM levels were particularly pronounced for patients requiring ventilatory support, those with septic shock, and those who died in the ICU, highlighting the potential of QFM for early severity assessment and mortality-risk predictions. Our findings provide a wider scope of QFM applications to critically ill populations with respiratory diseases beyond patients with COVID-19, suggesting its value in ICU settings. Further research is needed to explore the prognostic utility of QFM analysis and refine its clinical applications for managing immune dysfunction in critically ill patients.
XML Download