|Year : 2021 | Volume
| Issue : 1 | Page : 20-26
Can serum ferritin be employed as prognostic marker of pediatric septic shock and severe sepsis?
Mihir Sarkar1, Satyabrata Roychowdhury1, Mohammad Asraf Uz Zaman2, Sumantra Raut2, Subhajit Bhakta2, Mousumi Nandy2
1 Department of Pediatrics, PICU Unit, Medical College and Hospital, Kolkata, West Bengal, India
2 Department of Pediatrics, Medical College and Hospital, Kolkata, West Bengal, India
|Date of Submission||15-Jul-2020|
|Date of Decision||27-Sep-2020|
|Date of Acceptance||13-Oct-2020|
|Date of Web Publication||08-Jan-2021|
Dr. Mihir Sarkar
Department of Pediatrics, Medical College and Hospital, 88, College Street, Kolkata - 700 073, West Bengal
Source of Support: None, Conflict of Interest: None
Introduction: Serum ferritin has emerged as an independent marker predicting the outcome of critically ill patients. The cutoff value of ferritin in predicting mortality in septic children is not well established.
Objective: The aim is to determine the cutoff value of ferritin which is, best predictive of mortality in patients with age from 1 month to 12 years with a diagnosis of septic shock or severe sepsis and to evaluate correlation with PRISM III score, and Pediatric Logistic Organ Dysfunction score (PELODS).
Materials and Methods: A prospective observational study was conducted over a period of 2 years at pediatric intensive care unit (PICU) in children with age from 1 month to 12 years with septic shock or severe sepsis and PICU stay >24 h. Children with autoimmune diseases, primary hemophagocytic lymphohistiocytosis, blood transfusion in the past 4 months, known malignancies and immunosuppressive treatment were excluded. PRISM-III and PELODS were assessed to predict the risk of mortality and severity of disease. The highest value of ferritin was taken to draw area under the curve using receiver operating characteristic curve and determine cutoff value.
Results: Out of 176 children of septic shock or severe sepsis, 132 were included in the study. Mortality rate was 22.7% (n = 30). PRISM III and PELODS-2 were significantly high in nonsurvivors (P ≤ 0.001 and 0.006, respectively). The cutoff value of ferritin at 2375 ng/dl had sensitivity 96.7% and specificity 88% to predict mortality. Ferritin level was positively correlated to the PRISM III and PELODS, correlation coefficient 0.447 and 0.601, respectively.
Conclusion: Serum ferritin values ≥2375 ng/mL in children with septic shock, and severe sepsis was significantly associated with mortality. Its performance had a good correlation with PRISM III and PELODS.
Keywords: Biomarker, hyperferritinemia, mortality, septic shock
|How to cite this article:|
Sarkar M, Roychowdhury S, Uz Zaman MA, Raut S, Bhakta S, Nandy M. Can serum ferritin be employed as prognostic marker of pediatric septic shock and severe sepsis?. J Pediatr Crit Care 2021;8:20-6
|How to cite this URL:|
Sarkar M, Roychowdhury S, Uz Zaman MA, Raut S, Bhakta S, Nandy M. Can serum ferritin be employed as prognostic marker of pediatric septic shock and severe sepsis?. J Pediatr Crit Care [serial online] 2021 [cited 2021 Jan 26];8:20-6. Available from: http://www.jpcc.org.in/text.asp?2021/8/1/20/306478
| Introduction|| |
Sepsis and septic shock are major health-care problems, affecting millions of children around the world each year. Globally, an estimated 22 cases of childhood sepsis diagnosed per 100,000 person-years translating into 1.2 million cases of childhood sepsis per year. Mortality for children with sepsis ranges from 4% to as high as 50%, depending on illness severity, risk factors, and geographic location.
Septic shock is a complex syndrome displaying a tremendous degree of heterogeneity. Mortality in sepsis is mainly due from refractory shock and/or life-threatening organ dysfunction syndrome due to dysregulated host immune response to infection. Categorize sepsis subtypes through stratification of myriad biomarkers that are products of dysregulated inflammatory response may be fruitful endeavor with clinical implication.
Elevate serum ferritin is associated with several inflammatory conditions, such as sepsis, multiorgan dysfunction syndrome (MODS), and Macrophage Activation Syndrome. In addition to iron, ferritin synthesis is regulated by cytokines at various levels (transcriptional, posttranscriptional, and translational) during development, cellular differentiation, proliferation and inflammation., Ruddell et al. proposed extracellular ferritin can act as a pro-inflammatory signaling molecule in hepatic stellate cells which is entirely independent of its classic role as an iron-binding protein. Hence, serum ferritin has emerged as an independent marker of the severity of the disease and help in predicting the outcome of the critically ill patients. Garcia et al. from Brazil have described that elevated serum ferritin is often found in septic patients, and is related to worse clinical outcome in children.
The objective of this study was to determine the cutoff value of ferritin which is, best predictive of mortality in patients with age from 1 month to 12 years of age with a diagnosis of septic shock or severe sepsis and to evaluate correlation with PRISM III score, and Pediatric Logistic Organ Dysfunction score (PELOD).
| Materials and Methods|| |
This was a prospective single-centre observational study conducted over a period of 2 years (Feb 2018–Jan 2020) in children 1 month to 12 years with septic shock, severe sepsis and the multiple-organ dysfunction (MODS in pediatric intensive care unit (PICU) or high dependency unit of a tertiary care teaching hospital in eastern India. The study was approved by the Institutional ethics committee (approval no-MC/KOL/IEC/Non-Spon/707/12-2017), and children were enrolled in the study after parents gave informed consent.
Patients with sepsis, fulfilling inclusion criteria of age >1 month to <12 years and PICU stay more than 24 h were included in the study were included in the study. Exclusion criteria included: (1) children dying within 24 h of admission, (2) autoimmune diseases, (3) previous diagnosis of primary hemophagocytic lymphohistiocytosis (HLH) syndrome, (4) recipient of blood transfusion in the past 4 months, (5) children with known malignancies and immunosuppressive treatment. All consecutive patients meeting these criteria were included in the study.
Definition of sepsis
The diagnosis of severe sepsis and septic shock was based on the International pediatric sepsis consensus conference: Definitions for sepsis and organ dysfunction in pediatrics.
SIRS: Two of 4 criteria on core temperature, heart rate, respiratory rate, and leukocyte count. One of which must be abnormal temperature or abnormal leukocyte count.
Sepsis: SIRS in the presence of or as a result of suspected or proven infection.
Severe sepsis: Sepsis plus one of the following: cardiovascular organ dysfunction OR acute respiratory distress syndrome OR two or more other organ dysfunctions.
Septic shock:Sepsis plus cardiovascular dysfunction.
MODS: MODS is defined as a clinical syndrome characterized by the development of progressive and potentially reversible physiologic dysfunction in 2 or more organs or organ systems that is induced by a variety of acute insults.
The organ dysfunctions were classified according to Goldstein et al. The presence of two or more organic dysfunctions was considered as the MODS.
Following demographic and clinical variables were collected for the eligible enrolled patients: age, gender, nutritional status (malnutrition – weight-for-height ≤−2SD Z score), organ involvement, underlying disease, consanguinity, family history of HLH, duration of mechanical ventilation (MV), other respiratory support, amount of fluid bolus in first 6 h, need of renal replacement therapy (RRT), length of stay in the PICU and hospital along with outcome (survival/mortality). For the maximum vasoactive inotropic score (VIS), the highest value, obtained on any day of the study, was calculated through a summation obtained from the formula: dose of dopamine + dobutamine + (epinephrine × 100) + (noradrenaline × 100) + (vasopressin x 1000) + (milrinone × 10). All of them were expressed in mcg/kg/min except vasopressin, which was unit/kg/min.
Pediatric risk of mortality score (PRISM-III) was assessed to predict risk of death on 1st day of admission and PELODS-2 on day 1, 2, 5, 8, 12 were used to assess the severity of illness and organ dysfunction. Highest score was taken for statistical analysis. Blood culture findings, white blood cell count, hemoglobin (Hb) level, platelet count, alanine transaminase (ALT), aspartate transaminase, coagulation profile, C-reactive protein (CRP), and procalcitonin (PCT) were also recorded. Serum lactate from the time of admission (hyperlactatemia was defined as arterial lactate value of ≥2 mmol/L. Lactate was done at presentation then at 3 h then at 6 h interval till lactate level reduced to 2 mmol/L
Serum ferritin level of patients was measured on Day 1 and Day 3 of admission. Maximum value was taken for statistical analysis. Day 1 value was taken for analysis if patient died within 72 h. Ferritin measurement was done by Enzyme linked immunosorbent assay (ELISA) method (Dimension® EXL™ 200 Chemistry System, SIEMENS Healthcare) in our study.
The upper limit of normal ferritin used in boys aged 1–6 months, 7–12 months, 1–5 years, 6–9 years and 10–14 years was 410, 80, 24, 55 and 70 ng/mL, respectively. The upper limit of normal ferritin used in girls aged 1–6 months, 7–12 months, 1–5 years, 6–9 years and 10–14 years was 340, 45, 24, 55, and 40 ng/mL, respectively. We planned to determine the cutoff value by the area under the curve (AUC) in the receiver operating characteristic (ROC) curve.
To determine the serum ferritin cutoff value which is, best predictive of mortality in patients with age from Z to 12 years with the diagnosis of septic shock and severe sepsis.
Secondary outcome measures
(1) Evaluate correlation of Ferritin level with Pediatric risk of mortality score (PRISM III), PELODS-2, and (2) Investigate any significant difference in duration of MV, length of PICU stay, VIS, Lactate clearance time and Length of Hospital stay among patients dichotomized by the cutoff ferritin level.
Sample size was calculated assuming 80% of power of the study and α error of 0.05, high accuracy of ROC Curve at area of 0.9, and mortality rate 25% in severe sepsis in our center in the last 1 year, calculated sample size was 128.
Sample size was calculated with the help of SPSS Statistical Packages for the Social Sciences (SPSS version 23)(IBM SPSS Statistics for Windows, Version 23.0. Armonk, NY: IBM Corp). The same software was used for statistical analysis of data of this study. Patients were grouped according to mortality and clinical and laboratory parameters were compared in between groups. Categorical data were presented as count and percentage. Qualitative data were summarized as mean, median, standard deviation, and range. Quantitative continuous variables were compared between groups using Mann Whitney's non parametric tests if the variable had a nonnormal distribution or unpaired Student's t-test if the variable had a normal distribution. Qualitative variables were compared using Chi-square test (χ2) and Fisher's exact test. Correlation between variable were tested by Spearman rank correlation test and Correlation coefficient was calculated. ROC curve was analyzed to assess the predictability of the Ferritin level to discriminate between survivors and nonsurvivors. Youden index was calculated to determine the optimal cut-off points of Ferritin level. The AUC ranges between 0 and 1, and the proximity of AUC to 1 was considered as better performance of that variable. Secondary outcome parameters were compared in patients grouped by cutoff ferritin level. A P < 0.05 was considered statistically significant.
| Results|| |
Out of a total of 1220 admissions during the study period, 176 children (14.4%) had septic shock or severe sepsis [Figure 1]. Of these, 44 were excluded as prespecified exclusion criteria. Hence, the study included 132 patients. The overall in-hospital mortality rate of the study population was 22.7% (n = 30). At the time of presentation, 65.15% (n = 86) had septic shock, 22.7% (n = 30) had MODS and 12.15% (n = 16) had severe sepsis.
The distribution of baseline demographic and clinical parameters among Survivors and Nonsurvivors were described in [Table 1]. There were no differences in age and sex ratio in two groups. The incidence of malnutrition was similar in both the groups 15% versus 23% odds ratio 0.62 (P = 0.44). Duration of fever before admission was similar in both groups. Pneumonia was the most common source of infection. There were differences between survivors and nonsurvivors regarding presence of MODS and need of RRT with odds ratio 14.29 and 5.88, respectively, and P < 0.001 for both. While comparing PRISM III Score at 24 h nonsurvivors (19.9 ± 3.76) versus survivors (11.62 ± 2.79) had significantly (P ≤ 0.001) high score. Maximum PELODS-2 was more in nonsurvivors (P = 0.006). MV required 79 (68.6%) patients in survivors and 22 (73.3%) in nonsurvivors on day 1. Total 67 patients required >40 ml/kg fluid bolus on day 1 and mean VIS was (16.2 ± 1.8) in the study population. There was no significant difference of these two parameters among groups.
|Table 1: Demographic and clinical profile of patients with severe sepsis and septic shock|
Click here to view
Comparing laboratory values [Table 2] such as Hb, total leukocyte count, and platelet, coagulation profile, ALT, LDH, and Sodium value were not significantly different among the two groups (nonsurvivors and survivors). Blood culture was positive in 38.2% (n = 39) of survivors and 43.3% (n = 13) of nonsurvivors. Gram-negative organisms were most commonly isolated from blood culture. It was revealed that 24.5% (n = 25) of survivors and 26.6% (8) nonsurvivors grew Gram-negative organisms. Serum ferritin level was significantly high in nonsurvivors (7553 ± 2949 vs. 1293 ± 796, P = 0.001). While other biomarkers of sepsis in terms of CRP and procalcitonine were not significantly different in two groups, P = 0.628 and 0.08, respectively. Serum albumin (2.5 ± 0.39 vs. 3.15 ± 0.38, P = 0.001) was significantly low in nonsurvivors and creatinine value (1.4 ± 0.9 vs. 0.99 ± 0.7, P = 0.02) was significantly high. Significant high value of serum lactate at presentation was observed in nonsurvivors (4.6 ± 1.8 vs. 2.8 ± 0.9, P = 0.03). Abnormal radiological findings in chest X-ray did not differ statistically.
|Table 2: Comparison of laboratory parameters among survivors and nonsurvivors|
Click here to view
Ferritin cutoff value
The ROC curve analysis [Table 3], showed that the AUC of the ferritin for discrimination between survivors and nonsurvivors was 0.974, (95% confidence interval [CI] 0.952–0.996, P = 0.001). To achieve the maximum potential effectiveness of serum ferritin Youden index was applied. The cutoff value of ferritin at 2375 had sensitivity of 96.7% and specificity of 88% to predict mortality in the study population. Positive likelihood ratio was 7.04 (95% CI 4.31–12). It was compared to the AUC of the PRISM-III and PELODS [Figure 2]. In this subset of patients with septic shock and severe sepsis, serum ferritin level is comparable to PRISM III and PELODS to predict mortality. The cutoff value of PRISM III at 19.5 had 94% sensitivity and 97% specificity. Sensitivity and specificity of PELODS at cutoff value of 16.5 had 86.4% sensitivity and 89.2% specificity.
|Table 3: Receiver operating characteristic curve analysis for prediction mortality with the serum ferritin and other predictors of mortality|
Click here to view
|Figure 2: Analysis of area under the curve of receiver operating characteristic curves for prediction of mortality using the serum ferritin, PRISM III and Pediatric Logistic Organ Dysfunction Score|
Click here to view
We found that the ferritin level was positively correlated to the PRISM III and PELODS, Spearman Correlation coefficient 0.447 and 0.601, respectively, and P < 0.01 [Table 4]. Significant positive correlation was also found with serum creatinine level and need of renal replacement therapy (correlation coefficient 0.539 and 0.466, respectively. VIS had significant correlation with ferritin level. No significant correlation was noted with duration of MV and duration of PICU stay. Serum albumin level had negative correlation (r2 - 0.088) but was not significant statistically.
|Table 4: Spearman correlation between serum ferritin and other variables in the patients under the study|
Click here to view
It was observed that 42 patients in the study population had serum ferritin level >2375 ng/ml and out of which 29 (69%) patients had mortality. PRISM III and PELODS were higher in patients with hyperferritinemia (18.25 ± 3.42 vs. 11.79 ± 3.18, P = 0.019 and 20.11 ± 5.03 vs. 12.67 ± 3.92, P = 0.006) [Table 5]. It was revealed that VIS and Lactate clearance time was significantly high (P < 0.001for both) in hyperferritinemia patients. Need of renal replacement therapy differed significantly in the groups (42.85% vs. 3.09%, P = 0.001). Duration of MV and PICU length of stay was not statistically significant between groups.
|Table 5: Comparison of outcome parameters of patients dichotomized by ferritin level ≥2375 ng/ml|
Click here to view
| Discussion|| |
Ferritin was the most prominent inflammatory marker in this study. High value of this marker represents an intense inflammatory response scenario that seems to be an indicator of unfavorable outcomes. The present study found that the cutoff value of Ferritin at 2375 ng/ml had a sensitivity of 96.7% and specificity of 88% to predict mortality in children with septic shock and severe sepsis. Mortality in our study population with septic shock and server sepsis was 22.7%, which is comparable to other similar studies.,
According to Bennett et al., in the pediatric population, serum ferritin levels of >3000 ng/mL, was associated with increased risk for both receipt of critical care and subsequent death. On the other hand, Garcia et al. have studied it in a small number of patients (n = 36) and reported 58% mortality in children with high ferritin levels >500 ng/dl and severe sepsis. In another pediatric study from India by Sharma and Sharma found that ferritin value >1100 ng/ml had a 58.9% sensitivity and 75.3% specificity to predict mortality with a relative risk of 2.38 (95% CI: 1.57-3.61) and an odds ratio of 4.36 (95% CI: 2.14–8.88). The area under ROC curve was 0.68. In an adult study done by Lachmann et al. found that in septic shock maximum ferritin level was 1545 ng/ml. We observed that, serum ferritin cutoff value at 2375 ng/ml was a good predictor of mortality in children with septic shock and severe sepsis.
Several methods are in place to determine Serum Ferritin like-EIA: Enzyme immunoassay; ELISA: Enzyme linked immunosorbent assay; MEIA: Microparticle enzyme immunoassay; RIA: Radioimmune assay; IRMA: Immunoradiometric assay; LPIA Latex photometric immunoassay; RPIA: Radial partition immunoassay. Garcia-Casal et al. in their meta-analysis found good correlation (r = 0.878, 95% CI 0.869–0.886) between different methods. The methods used to determine ferritin are comparable and there is not preferred/recommended laboratory method. In this study, we used ELISA method (Dimension® EXL™ 200 Chemistry System, SIEMENS Healthcare).
It is evident from our study that high values of PRISM and PELODS score have a poor outcome. There were good correlation of Ferritin level with PELODS and PRISM III score-(Spearman correlation coefficient 0.601 and 0.447, respectively). AUC of the PRISM-III, PELODS and Ferritin was comparable also. Sharma and Sharma found that PRISM and PELODS were independently associated with mortality and also had a weak correlation with serum ferritin value. Contrary to our study, Ferritin was found to have weak correlation with PELODS (r2 = 0.23; P < 0.01) by Garcia et al.
Higher creatinine value and need of renal replacement therapy was associated with higher mortality. There was moderate correlation between ferritin level, need for RRT and serum creatinine level (Spearman correlation coefficient 0.446 and 0.539, respectively). Conflicting result was observed by Dimitrijevic et al. which showed that serum ferritin levels on admission may be used as a prognostic marker for predicting renal recovery in AKI patients, although they excluded patients with sepsis and infection.
The other two established inflammatory markers such as CRP and PCT were not significantly different among survivors and nonsurvivors in the present study. Another pediatric study also found no association was seen between CRP and ferritin value (r = 0.160), PCT had a weak direct correlation with serum ferritin values (0.230). CRP and Procalcitonine were not significantly associated with increased mortality in our study population. However, Carcillo et al. found that PICU mortality was increased in high-risk CRP ≥ 4.08 mg/dl category compare to CRP less that 4.08 mg/dl group.
Ghosh et al. in their study population, who had anemia in 85% of patients, observed that ferritin level in nonsurvivors 763 ng/mL (480–1820 ng/mL), whereas 415 ng/mL (262–852 ng/mL) in survivors (P = 0.11). This was much lower than what we revealed in our study. This may be explained by the fact our study group had anemia in total 17 (12.8%) patients. This suggests that in anemic children the cutoff of ferritin level may be lower to predict outcome.
Duration of MV and length of PICU stay did not differ significantly in patients stratified by ferritin level 2375 ng/ml. This may be due to high and early mortality in patients with hyperferritinemia.
There are some limitations in our study. First, it is a single-center study. A multicenter study with a bigger sample size and more power of the study could have predicted the cutoff level of ferritin more precisely to predict outcome in children with septic shock and severe sepsis. Second, we had not excluded patients with sepsis induced hepatic dysfunction which may change Ferritin metabolism and can affect study results. Third, serum ferritin is a reflection of iron stores in healthy individuals. Investigations to estimate body iron store could not be done.
| Conclusion|| |
Serum ferritin values ≥2375 ng/mL in pediatric patients with septic shock and severe sepsis admitted to the PICU, was significantly associated with mortality. Its performance had a good correlation with PRISM III and PELODS. Hyperferritinemia can be a useful prognostic biomarker for septic shock. Prospective multicenter studies are required for better assessment of the cutoff value of Ferritin.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Bryce J, Boschi-Pinto C, Shibuya K, Black RE; WHO Child Health Epidemiology Reference Group. WHO estimates of the causes of death in children. Lancet 2005;365:1147-52.
Fleischmann-Struzek C, Goldfarb DM, Schlattmann P, Schlapbach LJ, Reinhart K, Kissoon N. The global burden of pediatric and neonatal sepsis: A systematic review. Lancet Respir Med 2018;6:223-30.
Odetola FO, Gebremariam A, Freed GL. Patient and hospital correlates of clinical outcomes and resource utilization in severe pediatric sepsis. Pediatrics 2007;119:487-94.
Marshall JC, Reinhart K; International Sepsis Forum. Biomarkers of sepsis. Crit Care Med 2009;37:2290-8.
Demirkol D, Yildizdas D, Bayrakci B, Turkish secondary HLH/MAS critical care study Group: Hyperferritinemia in the critically ill child with secondary hemophagocytic lymphohistiocytosis/sepsis/multiple organ dysfunction syndrome/macrophage activation syndrome: what is the treatment? Crit Care 2012;16:R52.
Torti FM, Torti SV. Regulation of ferritin genes and protein. Blood 2002;99:3505-16.
Ruddell RG, Hoang-Le D, Barwood JM, Rutherford PS, Piva TJ, Watters DJ, et al
. Ferritin functions as a proinflammatory cytokine via iron-independent protein kinase C zeta/nuclear factor kappaB-regulated signaling in rat hepatic stellate cells. Hepatology 2009;49:887-900.
Gabay C, Kushner I. Acute phase proteins and other systemic responses to inflammation. New England J Med 1999;340:448-54.
Garcia PC, Longhi F, Branco RG, Piva JP, Lacks D, Tasker RC. Ferritin levels in children with severe sepsis and septic shock. Acta Paediatr 2007;96:1829-31.
Goldstein B, Giroir B, Randolph A. And the members of the international consensus conference on pediatric sepsis: international pediatric sepsis consensus conference: Definitions for sepsis and organ dysfunction in pediatrics. Pediatr Crit Care Med 2005;6:2-8.
Gaies MG, Gurney JG, Yen AH, Napoli ML, Gajarski RJ, Ohye RG, et al
. Vasoactive-inotropic score as a predictor of morbidity and mortality in infants after cardiopulmonary bypass. Pediatr Crit Care Med 2010;11:234-8.
Pollack MM, Patel KM, Ruttimann UE. PRISM III: An updated pediatric risk of mortality score. Crit Care Med 1996;24:743-52.
Leteurtre S, Duhamel A, Salleron J. Groupe Francophone de Réanimation et d'Urgences Pédiatriques (GFRUP). PELOD2: An update of the PEdiatric logistic organ dysfunction score. Crit Care Med 2013;41:1761-73.
Brugnara C. Reference values in infancy and childhood. In: Nathan DG, Orkin SH, editors. Nathan and Oski's hematology of infancy and childhood. 5th
ed. Philadelphia, PA: W.B. Saunders Company, 1998: Xxii.
Sankar J, Sankar MJ, Suresh CP, Dubey NK, Singh A. Early goal-directed therapy in pediatric septic shock: Comparison of outcomes “with” and “without” intermittent superior venacaval oxygen saturation monitoring: A prospective cohort study*. Pediatr Crit Care Med 2014;15:e157-67.
Bennett TD, Hayward KN, Farris RW, Ringold S, Wallace CA, Brogan TV. Very high serum ferritin levels are associated with increased mortality and critical care in pediatric patients. Pediatr Crit Care Med 2011;12:e233-6.
Sharma J, Sharma R. Serrum ferritin: A prognostic marker in patients with sepsis in Pediatric age group: A prospective cohort study. Int J Med Health Res 2018;4:86-89.
Lachmann G, Knaak C, Vorderwülbecke G, La Rosée P, Balzer F, Schenk T, et al. Spies C. Hyperferritinemia in Critically Ill Patients. Crit Care Med. 2020;48:459-65.
Garcia Casal MN, Peña Rosas JP, Urrechaga E, Jesus F. Escanero, Junsheng Huo, Ricardo X. Martinez, et al. Performance and comparability of laboratory methods for measuring ferritin concentrations in human serum or plasma: A systematic review and meta analysis. PLoS One 2018;13:E0196576.
Dimitrijevic ZM, Salinger-Martinovic SS, Jankovic RJ, Mitic BP. Elevated serum ferritin levels are predictive of renal function recovery among patients with acute kidney injury. Tohoku J Exp Med 2019;248:63-71.
Carcillo JA, Sward K, Halstead ES, Telford R, Jimenez-Bacardi A, Shakoory B, et al
. A systemic inflammation mortality risk assessment contingency table for severe sepsis. Pediatr Crit Care Med 2017;18:143-50.
Ghosh S, Baranwal AK, Bhatia P, Nallasamy K. Suspecting hyperferritinemic sepsis in iron-deficient population: Do we need a lower plasma ferritin threshold? Pediatr Crit Care Med 2018;19:e367-73.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]