Ferritin is an acute-phase reactant, commonly triggered by inflammation and infection. Accordingly, elevated serum ferritin is often found in septic patients, where it is related to worse clinical outcome (1,2). Other conditions associated with hyperferritinemia include chronic kidney disease, malignancies, particularly hemato-oncologic neoplasias, liver diseases, chronic alcohol consumption, and metabolic syndrome (3,4). Elevated ferritin is also seen in states of iron overload either due to blood transfusions or due to hemochromatosis, and in rather rare conditions such as hereditary hyperferritinemia and hemophagocytic lymphohistiocytosis (HLH) (4,5). The predictive value of ferritin as a potential marker for clinical outcome in the critically ill has been described by Bobbio-Pallavicini et al (6), who found an increase in ferritin to be associated with a deteriorating clinical condition. Further research conducted by Bennett et al (7) demonstrated that ferritin levels greater than 3,000 µg/L were related to ICU admission and increased mortality in pediatric patients, whereas Unal et al (8) identified elevated ferritin as an independent risk factor for death in adult ICU patients. Furthermore, a decrease in ferritin concentrations predicts clinical response in septic patients (9). In patients with HLH who commonly share clinical characteristics with septic patients, hyperferritinemia is one of eight diagnostic criteria according to the HLH-2004 criteria and is seen as an important diagnostic marker (10–13). Triggered by infections, malignancies, or autoimmune disorders, secondary HLH is characterized by toxic immune activation, leading to uncontrolled inflammation where extreme hyperferritinemia is observed. Due to a similar clinical presentation as sepsis, HLH in critically ill patients often remains undiagnosed, highlighting the relevance of accurate diagnostic markers (14). However, the optimal ferritin cutoff for HLH diagnosis in adult critically ill patients remains controversial, particularly to distinguish HLH from sepsis (15). An elevated ferritin of greater than 500 µg/L, as proposed by the Histiocyte Society, provides inadequate diagnostic specificity in adult HLH patients, as multiple conditions related to any type of inflammation can cause increased ferritin in the critical care setting (16). Small studies, determining prediction accuracy of ferritin for HLH diagnosis, revealed markedly higher thresholds. In adult ICU patients, our research group as well as Saeed et al (11) reported an optimal ferritin level of 3,095 µg/L and 3,951 µg/L, respectively (14). Barba et al (17) detected a median ferritin of 10,600 µg/L in their cohort of 71 adult HLH patients admitted to the ICU. Yet, concerns have been raised that even severely elevated ferritin is not as sufficiently pathognomonic for HLH as it is an unspecific marker of multiple processes including inflammation, iron overload, and renal and hepatic failure (9,15,18,19). Of note, Schram et al (18) found hemolytic anemia as the only condition to be independently associated with an increase in ferritin. As elevation of ferritin is commonly seen in the critical care setting and HLH is a rare diagnosis, its positive predictive value for HLH diagnosis is rather low (5,15).
Given the wide variety of etiologies underlying hyperferritinemia, the importance of ferritin in diagnosis of HLH, and the paucity of available data for ICU patients, our aim was to investigate the distribution of ferritin levels and associated conditions such as HLH, septic shock, sepsis, and other diagnoses in a large cohort of multidisciplinary ICU patients at an academic medical center to determine the ferritin cutoff best predictive for HLH in ICU.
This retrospective observational study was performed at the Charité – Universitätsmedizin Berlin. Data of ICU patients who were admitted to at least one of our adult surgical, anesthesiologic, or medical ICUs between January 2006 and August 2018 were extracted from two electronic patient data management systems operated at the Charité – Universitätsmedizin Berlin (COPRA, Sasbachwalden, Germany; and SAP, Walldorf, Germany). All patients 18 years old or older with at least one ICU ferritin value and hyperferritinemia of at least 500 µg/L in line with the HLH-2004 criteria (10) were included in the study. The study period was defined from admission to the ICU until hospital discharge, transfer, or death. As part of a post hoc analysis, we additionally searched all included patients for soluble interleukin 2 receptor (sIL-2R) as one of the other HLH-2004 criteria. In patients with multiple ferritin measurements, the highest ferritin value was considered for further analysis. The same applied to sIL-2R measurements.
Diagnosis of HLH, Sepsis, Septic Shock, and Other Diagnoses
As a first step, we searched the International Classification of Diseases, 10th Edition (ICD-10) codes (for HLH D76.1, D76.2, and D76.3) as well as the records of all patients for clinically diagnosed or suspected HLH, that were then reviewed by two HLH experts who confirmed or rejected the diagnosis. Only cases suspected or diagnosed with HLH by the clinicians were reviewed by the experts, which was according to the HLH-2004 criteria (10), at present the gold standard for HLH diagnosis, and the HScore (20) while taking into account patients’ history and clinical judgment (21). Of note, HLH patients include seven previously undiagnosed cases of HLH who have been retrospectively diagnosed and published by our research group (14). In a second step, we divided all non-HLH patients into three groups based on their ICD-10 codes: sepsis (A22.7, A39.1, A39.2, A39.3, A39.4, A40.-, A41.-, B37.7, O08.2, O75.3, O85, O88.3, R65.0, R65.1, T80.2, T81.4, T88.0; all without R57.2), septic shock (R57.2), and other diagnoses. In a third step, all patients with other diagnoses were allocated to 17 groups: liver disease, renal disease, autoimmune disease, hepatitis, tuberculosis, human immunodeficiency syndrome (HIV), Herpes simplex virus, Cytomegalovirus, Epstein-Barr virus, varicella-zoster virus (VZV), influenza, malaria, (bacterial/viral/fungal) infection, inflammation without infection, hematologic malignancy, solid tumors, and history of (stem cell/organ) transplantation (for ICD-10 codes of all groups, see Supplemental Table 1, Supplemental Digital Content 1, http://links.lww.com/CCM/F170).
Results are shown as median ± quartiles or percentage, respectively. Differences between patients with HLH, sepsis, septic shock and other diagnoses, respectively, were calculated using Kruskal-Wallis test for continuous data and the chi-square test for qualitative data. Receiver operating characteristics (ROC) analyses were performed to analyze best maximum ferritin levels and maximum sIL-2R to predict HLH and in-hospital mortality, respectively. As a post hoc analysis to test the predictive value of maximum ferritin to distinguish between two disease groups, we used repetitive multivariable logistic regression models adjusting for age, sex, body mass index (BMI), and maximum Sequential Organ Failure Assessment (SOFA) score with diagnoses as binary outcome variables (e.g., HLH vs septic shock, HLH vs sepsis, etc). In addition, ROC analyses were used to find the best maximum ferritin levels to differentiate between the respective cohorts. Multivariable logistic regression analysis was performed to analyze the influence of maximum ferritin on in-hospital mortality while adjusting for age, sex, BMI, diagnoses (categorical HLH/sepsis/septic shock/other), and maximum SOFA score. For all regression analyses, the natural logarithm of maximum ferritin values was used due to skewed data. SPSS 25.0 software (IBM, Armonk, NY) was used for all statistical analysis. A p value of less than 0.05 was considered statistically significant.
Ethics approval was obtained from the institutional review board (Ethikkommission der Charité – Universitätsmedizin Berlin, EA1/176/16). The study was registered with www.ClinicalTrials.gov (NCT02854943) on August 1, 2016.
Study Population and Outcome
In total, 116,310 patients were admitted to the ICUs between January 2006 and August 2018. Of these, 6,340 patients were 18 years old or older and had at least one ferritin measurement during their ICU stay. Ferritin was elevated (≥ 500 µg/L) in 2,623 patients, who were, therefore, finally analyzed (Supplemental Fig. 1, Supplemental Digital Content 2, http://links.lww.com/CCM/F171; legend, Supplemental Digital Content 6, http://links.lww.com/CCM/F175). Fifty of 2,623 patients had initially been diagnosed or suspected as HLH, of whom 40 were confirmed as HLH by the experts (n = 40 [1.52%]). For diagnosis of HLH, median 5 of 8 HLH-2004 criteria were fulfilled (minimum–maximum range, 3–7), whereas median HScore was 224 (minimum–maximum range, 133–302). For detailed description of all 40 HLH patients (Knaak et al ). Of the remaining 2,583 patients, 1,003 had sepsis without shock, 626 were diagnosed with septic shock, and 954 had other diagnoses. Five hundred ninety-six patients had more than one ferritin measurement (22.7%). Figure 1 depicts the distribution of all maximum ferritin levels in relation to patient numbers. Basic patient characteristics, biomarkers, and outcome parameters are shown in Table 1.
Serum Ferritin in HLH, Sepsis, Septic Shock, and Other Diagnoses
Over all groups, maximum ferritin levels differed significantly between all major patient groups (p < 0.001) (Fig. 2). Maximum ferritin levels were higher in HLH patients compared with patients with septic shock, sepsis, or other diagnosis (each p < 0.001). In sepsis and septic shock patients, maximum ferritin levels were higher than in patients with other diagnoses (each p < 0.001). When comparing sepsis and septic shock patients, maximum ferritin levels were higher in the latter group (median, 1,448 vs 1,545 µg/L; p = 0.001). Considering only patients with other diagnoses, Table 2 shows the median of the patients’ maximum ferritin values in various disease entities, with highest ferritin levels seen in patients with VZV, hepatitis, or malaria. In multivariable logistic regression analyses, ferritin was found as a good marker to distinguish HLH from septic shock, sepsis, and other diseases, but as poor marker to differentiate septic shock from sepsis or other diseases (Supplemental Table 2, Supplemental Digital Content 3, http://links.lww.com/CCM/F172).
Ferritin and sIL-2R for Diagnosis of HLH
ROC analysis identified a ferritin of 9,083 µg/L as predictive for HLH with 92.5% sensitivity and 91.9% specificity (AUC, 0.963; 95% CI, 0.949–0.978; Supplemental Fig. 2, Supplemental Digital Content 4, http://links.lww.com/CCM/F173; legend, Supplemental Digital Content 6, http://links.lww.com/CCM/F175). Sensitivity and specificity of various thresholds are shown in Supplemental Table 3 (Supplemental Digital Content 5, http://links.lww.com/CCM/F174). For all 94 available sIL-2R values, ROC analysis showed a value of 4,621 U/mL to be predictive for HLH with 66.7% sensitivity and 76.6% specificity (AUC, 0.752; 95% CI, 0.650–0.853; Supplemental Fig. 2, Supplemental Digital Content 4, http://links.lww.com/CCM/F173; legend, Supplemental Digital Content 6, http://links.lww.com/CCM/F175).
Ferritin as Prediction Marker of Mortality
Multivariable regression analysis including age, sex, BMI, maximum SOFA score, and diagnosis as confounders revealed statistically significant associations between maximum ferritin concentrations and in-hospital mortality (Table 3). The ROC curve showed only low discrimination ability for maximum ferritin on in-hospital mortality as measured by the AUC of 0.655 (95% CI, 0.631–0.679).
This is the first study describing the clinical characteristics of severe hyperferritinemia in a mixed ICU cohort, including 40 HLH cases. Furthermore, it is currently the largest report of ICU patients with ferritin levels available who had a broad spectrum of underlying diseases. We observed statistically significant differences in ferritin levels among patients with HLH, sepsis, septic shock, and other disease states. Among these groups, HLH patients had the highest ferritin levels; a cutoff of 9,083 μg/L revealed the highest sensitivity (92.5%) and specificity (91.9%) for the diagnosis of HLH. Comparing sepsis and septic shock patients only, ferritin values were higher in septic shock patients. Analyzing conditions associated with hyperferritinemia other than HLH, sepsis, and septic shock, those with highest ferritin levels were VZV, hepatitis, and malaria.
Hyperferritinemia is frequently seen in the ICU in response to malignant, liver, renal, or inflammatory disorders (1,5,6,11,15). Among the latter, sepsis is one of the most common causes. However, also HLH should be considered as a differential diagnosis, albeit a rare one, especially in the case of nonresponsiveness to anti-infective treatment, cytopenias, prolonged fevers, or organomegaly. Overlap in clinical features is challenging in differentiating HLH from sepsis. Fever, cytopenias, and hyperferritinemia can be present in both. However, those findings do not confirm nor rule out either HLH or sepsis. According to our findings and previous studies by others (11,23), there is now evidence for a higher threshold of ferritin to diagnose secondary HLH in the ICU. In our cohort, highest ferritin levels were seen in HLH patients, followed by septic shock and sepsis patients, surpassing the levels in all other diagnoses. However, despite these encouraging results for identifying HLH, ferritin must always be interpreted in the context of the patient’s underlying condition, clinical course, and additional HLH diagnostic criteria. As argued by Sackett et al (5), ferritin has a rather low positive predictive value, possibly placing patients at risk to overtreatment if HLH is considered but in fact not present. In addition, our data revealed that severe hyperferritinemia is present in multiple other conditions, including tuberculosis, hepatitis, and other viral infections. Nevertheless, ferritin was still highest in HLH compared with all other conditions. These findings have implications for clinical practice. Differential diagnosis of the causes of hyperferritinemia will have to largely rely on the degree of ferritin elevation. The presence of severely elevated ferritin levels exceeding limits seen in tuberculosis, viral infections, renal and hepatic disorders, then warrants further diagnostics to rule out or to confirm a diagnosis of HLH. There is strong evidence that HLH is indeed an underdiagnosed condition (14), that is why we believe that early assessment of ferritin could increase detection rates of HLH, facilitate early treatment, and thereby contribute to improved patient outcome.
Based on our data, a ferritin of 9,083 μg/L was most predictive for HLH. This constitutes a considerably higher threshold than the cutoff of 500 µg/L as proposed in the HLH-2004 criteria. In their cohort of critically ill adult patients, Saeed et al (11) found an optimal cutoff at 3,951 µg/L. Similarly, previous research in our study group had revealed the best prediction accuracy at 3,095 µg/L (14), whereas in another study, in ICU patients (23), ferritin of 1,197 µg/L was correlated best with secondary HLH. In our analyses, even ferritin of 4,006 µg/L showed good prediction accuracy with 95% sensitivity and 83% specificity. However, it has been argued that ferritin alone provides insufficient specificity for diagnosing HLH (24). At a value of ferritin greater than 50,000 µg/L, Schram et al (18) found an only low sensitivity (< 20%) and specificity (17%) for HLH. The authors concluded that in view of the rarity of HLH, more common conditions also associated with hyperferritinemia such as renal and hepatic failure, infections, and malignancies should be considered first. Small sample size and heterogeneity in underlying conditions associated with HLH were potential limitations in previous studies. The studies by Saeed et al (11) and us (14) included nine patients each, most of whom developed septic shock during their clinical course. As our study comprised the currently largest mixed ICU cohort with ferritin available, we could take into account the wide variety of hyperferritinemia-related conditions encountered in the critical care setting and believe that a concentration of 9,083 µg/L for ferritin is a reliable marker, warranting further diagnostics to confirm or exclude HLH. Of note, the threshold of 500 µg/L was established for a pediatric population presenting with primary HLH (10). In this particular patient group, the cutoff of 500 µg/L is well justified because these patients might initially have ferritin levels considerably lower than 9,083 μg/L with a steep increase as HLH progresses. Those patients might remain undetected if the diagnostic threshold is raised without taking the possibility of primary HLH into account. However, in the adult ICU population, ferritin of 500 µg/L has only poor specificity in diagnosing HLH. We, therefore, believe that a markedly higher threshold should be used to avoid confounding HLH with other conditions associated with severe hyperferritinemia in the critical care setting.
Concerns have been raised that secondary HLH remains frequently unrecognized, leading to delayed or no treatment and subsequent high mortality rates, particularly in the critically ill (14,25–27). The frequency of 1.52% among ICU patients with ferritin of at least 500 µg/L found in our cohort warrants increased awareness of this life-threatening condition. Diagnosis of HLH is often hampered and delayed by its clinical overlap with sepsis or septic shock. As our data demonstrate, ferritin concentrations differ significantly among sepsis, septic shock, and HLH underscoring the role of highly elevated ferritin in recognition of HLH. Therefore, a sepsis-like condition in the presence of refractory fever and unexplained cytopenia demands testing of ferritin levels and further HLH diagnostics.
As one of the HLH-2004 diagnostic criteria, sIL-2R is also an important disease marker in HLH. A value of greater than or equal to 2,400 U/mL has a reported sensitivity and specificity of 93% and 100%, respectively, in pediatric patients (10). Zhang et al (28) reported sIL-2R as an independent marker to predict survival and monitor treatment response. Hayden et al (29) demonstrated an excellent predictive value at 2,515 U/mL with 100% sensitivity and 72.5% specificity in adult non-ICU HLH patients. However, we detected rather unsatisfying sensitivity (66.7%) and specificity (76.6%) at a value of 4,621 U/mL in our large mixed ICU cohort. Therefore, sIL-2R may not be a strong marker of HLH within the ICU population, but this assessment is limited by the relatively small number of sIL-2R testing within this cohort. Of note, at high values exceeding the respective laboratory quantitation standard, sIL-2R and serum ferritin samples require dilution to yield exact results. This is important, as reports of measurements beyond a certain threshold (e.g., > 7,500 U/mL) do not allow for exact clinical evaluation.
We found an increased ferritin concentration to be independently associated with higher mortality. Yet, when trying to determine a cutoff for mortality prediction, discriminatory performance was rather low in ROC analysis. Therefore, ferritin may not be strongly related to in-hospital mortality. Previous research conducted in medical ICU and pediatric critical care patients consistently demonstrated ferritin to be an independent factor for increased mortality (7–9). For HLH patients, Grangé et al (30) showed that ferritin was the only independent predictor of mortality. In line, a decrease in ferritin during therapy was associated with lower mortality rates emphasizing the potential prognostic value of ferritin (31).
Our study has several limitations. First, it is retrospective in nature, limiting data availability to ferritin assessments during hospitalization to the ICU. Because our cohort comprises patients admitted to various ICUs where some might have been more likely than others to have ferritin measurement, there might be a selection bias toward specific disease groups. Second, there is variability in the timing that ferritin levels were obtained as some patients had ferritin measurement at admission, others later during their ICU stay. Third, there is a considerable likelihood that HLH cases in ICU remained undetected based on clinicians’ experience. Fourth, as HLH diagnosis partially relied on expert review, there might be a bias toward HLH diagnosis. Fifth, the overlap of patients between our chosen disease groups might distort results. Finally, our data are derived from an academic medical center, which might limit generalization for a community hospital.
This is the first study describing clinical characteristics of severe hyperferritinemia in a mixed ICU cohort including 40 HLH cases. Furthermore, it is the largest study of patients with ferritin available in critically ill patients. Ferritin concentrations differed significantly, depending on the underlying condition. Hyperferritinemia was most pronounced in patients with HLH, followed by septic shock, sepsis, and other disease entities. At a cutoff of 9,083 µg/L, prediction accuracy was at 92.5% sensitivity and 91.9% specificity to detect HLH. Ferritin is easy to assess and should therefore be measured in ICU patients with unclear inflammation to increase the detection rate of HLH.
We thank the Department of Cardiovascular Surgery, the Department of Surgery CCM/CVK, the Medical Department, Division of Nephrology and Internal Intensive Care Medicine, the Medical Department, Division of Infectiology and Pneumonology, the Medical Department, Division of Cardiology (CVK), the Department of Cardiology (CBF), the Department of Neurology with Experimental Neurology, and the Department of Anesthesiology and Operative Intensive Care Medicine (CBF) for being part of our study, providing the data and excellent collaboration.
1. Piagnerelli M, Cotton F, Herpain A, et al. Time course of iron metabolism in critically ill patients. Acta Clin Belg 2013; 68:22–27
2. Garcia PC, Longhi F, Branco RG, et al. Ferritin
levels in children with severe sepsis and septic shock. Acta Paediatr 2007; 96:1829–1831
3. Senjo H, Higuchi T, Okada S, et al. Hyperferritinemia
: Causes and significance in a general hospital. Hematology 2018; 23:817–822
4. Cullis JO, Fitzsimons EJ, Griffiths WJ, et al.; British Society for Haematology: Investigation and management of a raised serum ferritin
. Br J Haematol 2018; 181:331–340
5. Sackett K, Cunderlik M, Sahni N, et al. Extreme hyperferritinemia
: Causes and impact on diagnostic reasoning. Am J Clin Pathol 2016; 145:646–650
6. Bobbio-Pallavicini F, Verde G, Spriano P, et al. Body iron status in critically ill patients: Significance of serum ferritin
. Intensive Care Med 1989; 15:171–178
7. Bennett TD, Hayward KN, Farris RW, et al. Very high serum ferritin
levels are associated with increased mortality
and critical care in pediatric patients. Pediatr Crit Care Med 2011; 12:e233–e236
8. Unal AU, Kostek O, Takir M, et al. Prognosis of patients in a medical intensive care unit
. North Clin Istanb 2015; 2:189–195
9. Carcillo JA, Sward K, Halstead ES, et al.; Eunice Kennedy Shriver National Institute of Child Health and Human Development Collaborative Pediatric Critical Care Research Network Investigators: A systemic inflammation mortality
risk assessment contingency table for severe sepsis. Pediatr Crit Care Med 2017; 18:143–150
10. Henter JI, Horne A, Aricó M, et al. HLH-2004: Diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis
. Pediatr Blood Cancer 2007; 48:124–131
11. Saeed H, Woods RR, Lester J, et al. Evaluating the optimal serum ferritin
level to identify hemophagocytic lymphohistiocytosis
in the critical care setting. Int J Hematol 2015; 102:195–199
12. Allen CE, Yu X, Kozinetz CA, et al. Highly elevated ferritin
levels and the diagnosis of hemophagocytic lymphohistiocytosis
. Pediatr Blood Cancer 2008; 50:1227–1235
13. Machowicz R, Janka G, Wiktor-Jedrzejczak W. Similar but not the same: Differential diagnosis of HLH and sepsis. Crit Rev Oncol Hematol 2017; 114:1–12
14. Lachmann G, Spies C, Schenk T, et al. Hemophagocytic lymphohistiocytosis
: Potentially underdiagnosed in intensive care units. Shock 2018; 50:149–155
15. Castillo L, Carcillo J. Secondary hemophagocytic lymphohistiocytosis
and severe sepsis/systemic inflammatory response syndrome/multiorgan dysfunction syndrome/macrophage activation syndrome
share common intermediate phenotypes on a spectrum of inflammation. Pediatr Crit Care Med 2009; 10:387–392
16. Aad G, Abbott B, Abdallah J, et al.; ATLAS Collaboration: Search for dilepton resonances in pp collisions at √s=7 TeV with the ATLAS detector. Phys Rev Lett 2011; 107:272002
17. Barba T, Maucort-Boulch D, Iwaz J, et al. Hemophagocytic lymphohistiocytosis
in intensive care unit
: A 71-Case Strobe-Compliant Retrospective Study. Medicine (Baltimore) 2015; 94:e2318
18. Schram AM, Campigotto F, Mullally A, et al. Marked hyperferritinemia
does not predict for HLH in the adult population. Blood 2015; 125:1548–1552
19. Carcillo JA, Halstead ES, Hall MW, et al.; Eunice Kennedy Shriver National Institute of Child Health and Human Development Collaborative Pediatric Critical Care Research Network Investigators: Three hypothetical inflammation pathobiology phenotypes and pediatric sepsis-induced multiple organ failure outcome. Pediatr Crit Care Med 2017; 18:513–523
20. Fardet L, Galicier L, Lambotte O, et al. Development and validation of the HScore, a score for the diagnosis of reactive hemophagocytic syndrome
. Arthritis Rheumatol 2014; 66:2613–2620
21. La Rosée P, Horne A, Hines M, et al. Recommendations for the management of hemophagocytic lymphohistiocytosis
in adults. Blood 2019; 133:2465–2477
22. Knaak C, Schuster FS, Spies C, et al. Hemophagocytic lymphohistiocytosis
in critically ill patients. Shock 2019 Oct 15. [Epub ahead of print]
23. Meena NK, Sinokrot O, Duggal A, et al. The performance of diagnostic criteria for hemophagocytic lymphohistiocytosis
in critically ill patients. J Intensive Care Medi 2019:885066619837139
24. Nikiforow S, Berliner N. The unique aspects of presentation and diagnosis of hemophagocytic lymphohistiocytosis
in adults. Hematology Am Soc Hematol Educ Program 2015; 2015:183–189
25. Price B, Lines J, Lewis D, et al. Haemophagocytic lymphohistiocytosis: A fulminant syndrome associated with multiorgan failure and high mortality
that frequently masquerades as sepsis and shock. S Afr Med J 2014; 104:401–406
26. Raschke RA, Garcia-Orr R. Hemophagocytic lymphohistiocytosis
: A potentially underrecognized association with systemic inflammatory response syndrome, severe sepsis, and septic shock in adults. Chest 2011; 140:933–938
27. Okabe T, Shah G, Mendoza V, et al. What intensivists need to know about hemophagocytic syndrome
: An underrecognized cause of death in adult intensive care units. J Intensive Care Med 2012; 27:58–64
28. Zhang L, Zhang S, Xu J, et al. Significance of soluble interleukin-2 receptor in patients with hemophagocytic lymphohistiocytosis
. Leuk Lymphoma 2011; 52:1360–1362
29. Hayden A, Lin M, Park S, et al. Soluble interleukin-2 receptor is a sensitive diagnostic test in adult HLH. Blood Adv 2017; 1:2529–2534
30. Grangé S, Buchonnet G, Besnier E, et al. The use of ferritin
to identify critically ill patients with secondary hemophagocytic lymphohistiocytosis
. Crit Care Med 2016; 44:e1045–e1053
31. Lin TF, Ferlic-Stark LL, Allen CE, et al. Rate of decline of ferritin
in patients with hemophagocytic lymphohistiocytosis
as a prognostic variable for mortality
. Pediatr Blood Cancer 2011; 56:154–155