International Journal of Medical and Health Research

International Journal of Medical and Health Research

ISSN: 2454-9142
www.medicalsciencejournal.com
Volume 4; Issue 3; March 2018; Page No. 86-89

Serrum ferritin: A prognostic marker in patients with sepsis in Pediatric age group: A prospective cohort study
Dr. Jyoti Sharma1, Dr. Roop Sharma2
1
DNB Pediatrics, Institute of Child Health, Sir Ganga Ram Hospital, New Delhi, India
2
MBBS, MD (Pediatrics) Fellow, PICU, Sir Gangaram Hospital, New Delhi, India

Abstract
Objectives: To determine serum ferritin value, to study its correlation with PRISM/PELOD score and mortality in patients with
sepsis.
Methods: This Analytical Observational prospective cohort study was done in12-bedded Pediatric intensive care unit (PICU) of
Sir Ganga Ram Hospital, New Delhi. 149 patients with sepsis were included for the study with an inclusion criteria of age >28
days or <16 years and PICU stay >24 hours. Serum ferritin was collected at the time of diagnosis of sepsis. Patient demographics
were noted and Pediatric risk of mortality (PRISM) score and Pediatric logistics dysfunction (PELOD) score were calculated. All
patients were followed up throughout their hospital stay and outcome data (survival or mortality) was obtained.
Results: Of the 149 patients, 93 survived while 56 died in PICU. Non survivors had a significant higher median ferritin (1205.50
ng/ml) as compared to survivors (p < 0.001). At a cut off of 1100 ng/ml, serum ferritin was associated with a 2.3 (1.571-3.614)
relative risk (p < 0.0001) for predicting mortality in patients with sepsis. PRISM and PELOD score were independently associated
with mortality and also had a weak correlation with serum ferritin value. No significant difference was found in C- reactive protein
(CRP) as well as procalcitonin value in survivors and non survivors group.
Conclusion: High serum ferritin level is associated with poor outcome in patients of sepsis and can be used as a predictive marker
of mortality along with current prognostic scores.

Keywords: ferritin, biomarkers, sepsis, children, pediatric intensive care, prognostic scores

1. Introduction remains difficult to differentiate sepsis from other non

Sepsis is a leading cause of death in critically ill patients infectious causes of systemic inflammatory response
despite the use of modern antibiotics and resuscitation syndrome (SIRS), and there is a continuous search for better
therapies [1]. Diagnosis of sepsis in children is difficult in biomarkers of sepsis.
everyday practice for many reasons: the clinical signs and New research and novel understanding of the molecular basis
symptoms in children are very variable and nonspecific at the of the disease reveals an abundance of exciting new markers
start of the infection; microbiological culture results are that may be of utility in clinical practice. One such marker is
expected only after 48-72 hours and false negatives are serum ferritin which is an iron storage protein that sequesters
common. In 2002, an International Sepsis Consensus iron in the ferric (Fe3+) state. It is a complex of iso ferritins
Conference (ISCC) held in USA led to framing and adoption produced by the reticulo endothelial (RE) system. The RE
of specific clinical definitions for sepsis [2]. Early diagnosis system plays a critical role in iron metabolism by processing
and stratification of severity of sepsis is very important, hemoglobin from senescent red blood cells. Acute
increasing the possibilities of initiating timely and specific inflammation and infection induce the blockade of iron release
treatment [3, 4]. resulting in a decreased serum iron, a virulence factor for
Biomarkers can indicate the presence or absence or severity of many microorganisms. Elevated levels of serum ferritin, an
sepsis [5, 6]. They also have roles in prognostication, guiding acute-phase reactant, reflect the clinical response to deprive
antibiotic therapy, evaluating the response of therapy and microorganisms of serum iron [14, 16]. There have been studies
recovery from sepsis predicting sepsis complications and the in adults which describe association of ferritin and critical care
development of organ dysfunction [7]. During the last decade, outcome. Taking into account the new definition of sepsis in
measurement of C reactive protein (CRP), a good children and evaluating the need for effective and rapid
inflammatory marker, has been added to the set of laboratory (quantitative) indicator, we studied the co-relation
hematological tests (total leukocyte count, neutrophils, band of serum ferritin with the diagnosis and prognosis of sepsis in
form counts) that have long been used in clinical practice. children.
However, it does not have the specificity required to
distinguish viral from bacterial infections. The calcitonin pro- Materials and Methods
hormone procalcitonin (PCT) has also been used clinically; its This analytical observational prospective cohort study was
level is low in healthy individuals (< 0.5 ng/ml) [8, 10]. It has conducted over a period of 16 months (May 2012-August
been proposed as a more specific and better prognostic marker 2013) in the pediatric intensive care unit (PICU) of Sir Ganga
than CRP, although its value has also been challenged [11, 13]. It Ram Hospital, Rajinder Nagar, New Delhi which is a 650-

86
International Journal of Medical and Health Research

bedded tertiary care centre. Approval was taken from hospital survivors (p=0.001, p<0.001, respectively) but no correlation
ethics committee and written consent was obtained from of either was seen with the serum ferritin levels (p=0.630,
parents. p=0.798) (Table 1). Furthermore, 114 (76.5%) patients
Children were selected on the basis of sepsis definition required mechanical ventilation out of which 50 (33.5%) died.
defined by the International Pediatric Sepsis Consensus Total number of deaths was 56 (37.6%). We studied
Conference [2]. 149 patients with sepsis were included for the association between use of vasoactive drugs17 and ferritin and
study with an inclusion criteria of age >28 days to <16 years found it to be positive (r = 0.362, p < 0.001). Such association
and PICU stay >24 hours. Exclusion criteria included (1) has not been reported so far.
pediatric surgical, trauma and burn cases, (2) children dying
within 24hours of PICU admission, (3) autoimmune diseases, Age Distribution
(4) evidence of malaria, (5) diagnosis of hemophagocytic
70%
syndrome, (6) recipient of blood transfusion in the last 4
59.1%
months, (7) cases of hepatitis, (8) children with other causes 60%
of shock, not due to sepsis itself, e.g., cardiogenic,
50%
anaphylactic, and dengue shock, (9) children with known

% of cases
malignancies and immunosuppressive treatment. 40%
Following data was collected for the eligible enrolled patients:
30%
age, gender, primary organ involvement, duration of
mechanical ventilation, vasoactive ionotropic score [17], length 20%
12.8% 11.4%
of stay in the PICU and hospital along with final outcome 6.0% 8.1%
10%
(survival/mortality). Pediatric risk of mortality score (PRISM) 2.7%
and pediatric logistic organ dysfunction score (PELODS) were 0%
used to assess the severity of illness and organ dysfunction. 0 - 3 yrs 3 - 6 yrs 6 - 9 yrs 9 - 12 yrs 12 - 15 yrs >=15 yrs
Patients with sepsis, severe sepsis and septic shock were
identified using definitions provided by Goldstein et al. [18] In Fig 1: Age Distribution (n=149)
addition, the results of all the investigations necessary for
Table 1. Correlation between duration of PICU/hospital stay and
routine management sent by PICU team at admission and in serum ferritin level
the next 24 hours were recorded.
Serum ferritin level was measured (at the time of diagnosis of Ferritin
sepsis) by immunochemiluminiscent assay which is a PICU Stay (in days)
Pearson Correlation 0.040
competitive immunometric assay involving liquid phase coefficient(r) p-value 0.630
ligand-labelled protein binding. Pearson Correlation -0.021
Hospital Stay (in days)
Continuous variables were presented as mean  standard coefficient(r) p-value 0.798
deviation or medians (minimum-maximum) as per the
distribution. Categorical variables were expressed as The median serum ferritin value was 761 (178-1886.50)
frequencies (%). Differences between groups were assessed ng/ml. Ferritin above 1100 ng/mL had 58.9% sensitivity and
with Chi-square or Fisher’s exact test for categorical variables. 75.3% specificity to predict death, and was associated with a
Unpaired t test was used for comparison of continuous 2.3 (1.571-3.614) relative risk of death (p < 0.0001) with
variables between the two groups. For non parametric data, 0.685 area under ROC curve (Figure 2).
Mann Whitney U test was used. The area under the receiver The median values of PRISM score at 12 hours and at 24
operator characterstic (ROC) was used to assess the ability to hours were 11 (range: 5-17) and 10 (range: 5-16) respectively.
predict outcome. Multivariate logistic regression analysis was Median values of PELOD score on day 1 and day 2 were 21
also done to predict an outcome from a set of predictor (range: 11-31) and 21 (range: 10.50-31) respectively. We also
variables in univariate analysis. P<0.05 was considered tested the correlation of PRISM/PELODS with ferritin values
statistically significant. Statistical analysis was performed and found that ferritin had a direct (though weak) correlation
with SPSS version 17.0 program for Windows (SPSS Inc., with PRISM at 12 hours and 24 hours (r =0.349; p<0.001,
Chicago, IL, USA). r=0.311; p< 0.001 respectively) and with PELOD score on
day 1 (r = 0.249; p =0.002) and day 2 (r=0.269; p=0.001).
Results Higher PRISM (> 15 at 12 hours) and PELOD (> 21 on day 1)
149 children were enrolled in the study with a median age of scores were independently associated with increased mortality
1.8 (0.5- 6) years with maximum cases (59.1 %) below 3 years (p< 0.001, p=0.001, respectively).
of age (Figure 1). There were 104 (69.8%) boys. Respiratory Known inflammatory biomarkers namely CRP and
system was involved in majority (34.9 %) of the cases Procalcitonin (PCT) were also studied. While no association
followed by CNS (17.4%) and GIT (14.1%). Although, no was seen between CRP and ferritin value (r=0.160, p=0.243),
significant difference was seen in serum ferritin value in PCT had a weak direct correlation with serum ferritin values
relation to primary organ involved. The duration of PICU and (0.230, p=0.005). Contrary to earlier studies done on
total hospital stay was higher in survivors as compared to non biomarkers, CRP and procalcitonin were not significantly
associated with increased mortality in our study cases
(p=0.839, p=0.346, respectively).

87
International Journal of Medical and Health Research

References
1. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G,
Carcillo J, Pinsky MR. Epidemiology of severe sepsis in
the United States: analysis of incidence, outcome, and
associated costs of care. Crit Care Med, 2001; 29:1303-
1310.
2. Randolph AG. The purpose of the 1st international sepsis
forum on sepsis in infants and children. Pediatr Crit Care
Med, 2005; 6:S1-S2.
3. Kumar A, Roberts D, Wood KE, Light B, Parrillo JE,
Sharma S et al. Duration of hypotension before initiation
of effective antimicrobial therapy is the critical
determinant of survival in human septic shock. Crit Care
Med, 2006; 34:1589-1596.
4. Zambon M, Ceola M, Almeida-de-Castro R, Gullo A,
Vincent JL. Implementation of the Surviving Sepsis
Campaign guidelines for severe sepsis and septic shock:
we could go faster. J Crit Care, 2008; 23:455-460.
5. Biomarkers Definitions Working Group. Biomarkers and
surrogate endpoints: preferred definitions and conceptual
Fig 2: ROC curve for serum ferritin framework. Clin Pharmacol Ther, 2001; 69:89-95.
6. Marshall JC, Reinhart K. Biomarkers of sepsis. Crit Care
Discussion Med. 2009; 37:2290-2298.
Sepsis caused by infection remains a major cause of mortality 7. Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM,
and morbidity among children. Several inflammatory markers Jaeschke R. et al. Surviving Sepsis Campaign:
have failed to meet the requirements for early diagnosis and international guidelines for management of severe sepsis
prognostication of sepsis. This study evaluated the potential and septic shock. Crit Care Med, 2008; 36:296-327.
value of measuring ferritin in patients with clinically 8. Guven H, Altintop L, Baydin A. et al. Diagnostic value of
suspected and proven sepsis, correlated it with procalcitonin levels as early indicator of sepsis. Am J
PRISM/PELODS and outcome of patients and compared it Emerg Med. 2002; 20:202-6.
with established inflammatory markers like C- reactive protein 9. Casado Flores J, Blanco Quiros A, Asesnsio J, Arranz E,
(CRP) and procalcitonin (PCT). Garrote J, Nieto M. Serum procalcitonin in children with
Ferritin has been reported to be elevated in some previous suspected sepsis. A comparison with C reactive protein
reports from cases of sepsis in adults [19, 20]. Garcia et al have and neutrophil count. Pediatr Crit Care Med, 2003; 4:190-
studied it in small number of patients ( n=36) in pediatric age 5.
group21. We have found that ferritin value >1100ng/ml had a 10. Hathrill M, Tibby SM, Sykes K, Turner C, Murdoch IA.
58.9% sensitivity and 75.3 % specificity to predict mortality Diagnostic markers of infection: comparison of
with a relative risk of 2.38 (95% CI :1.57-3.61) and an odds procalcitonin with C reactive protein and leucocyte count.
ratio of 4.36 (95% CI:2.14 -8.88).The area under ROC curve Arch Dis Child, 1999; 81:417-21.
was 0.685. Our findings have been consisitent with earlier 11. Nakamura A, Wada H, Ikejiri M, Hatada T, Sakurai H,
studies in adults. We also found out that ferritin has direct Matsushima Y. et al. Efficacy of procalcitonin in the
correlation with procalcitonin value but none with CRP. This early diagnosis of bacterial infections in a critical care
association has not been reported before in children. It is unit. Shock, 2009; 31:591-93.
evident from our study that high values of PRISM and 12. .Luzzani A, Polati E, Dorizzi R, Rungatscher A, Pavan R,
PELODS score have a poor outcome. So, these prognostic Merlini A. Comparison of procalcitonin and C-reactive
scores can be used along with ferritin to predict mortality in protein as markers of sepsis. Crit Care Med, 2003;
sepsis. 31:1737-1741.
There are some limitations of our study. First, we studied only 13. Tang BM, Eslick GD, Craig JC, McLean AS. Accuracy
the cases with clinical sepsis and no controls were included in of procalcitonin for sepsis diagnosis in critically ill
our study. Second, we did not exclude cases who were patients: systematic review and meta analysis. Lancet
admitted/treated elsewhere before being referred to our Infect Dis. 2007; 7:210-217.
hospital. Third, other components of iron metabolism like 14. Cunha BA, Sachdev B, Canario D. Serum ferritin levels
lactoferrin, transferrin etc. which can affect iron availability, in West Nile encephalitis. Clin Microbiol Infect. 2004;
were not studied in our cases due to financial limitation. 10:184-6.
Despite of these limitations, ferritin appears to be an important 15. Fahmy M, Young SP. Modulation of iron metabolism in
early prognostic marker for mortality in sepsis, hence, should monocyte cell line U937 by inflammatory cytokines:
be included as a part of routine sepsis work up. Its predictive changes in transferrin uptake, iron handling and ferritin
performance can be increased with use of existing prognostic mRNA. Biochem J. 1993; 296:175-81.
scores. More studies are needed to study effect of deranged 16. Smirnov IM, Bailey K, Flowers CH, Garrigues NW,
iron profile on ferritin values. Wesselius LJ. Effects of TNF-alpha and IL-1 beta on iron

88
International Journal of Medical and Health Research

metabolism by A549 cells and influence on cytotoxicity.

Am J Physiol. 1999; 277:257-63.
17. Michael Gaies G, James Gurney G, Alberta Yen G,
Michelle Napoli L, Robert Gajarski J, Richard Ohye G, et
al. Vasoactive–inotropic score as a predictor of morbidity
and mortality in infants after cardiopulmonary bypass.
Pediatr Crit Care Med, 2010; 11:234-238.
18. Brahm Goldstei, Brett Giroir, Adrienne Randolph, and the
Members. International Consensus Conference on
Pediatric Sepsis, Pediatr Crit Care Med, 2005; 6(1):2-8.
19. Diaz J, Arribas JM, Vallina E, Maradona JA, Hevia C,
Blanco F. Acute-phase reactants in sepsis. Rev Clin Esp,
1992; 191:473-7.
20. Darveau M, Denault AY, Blais N, Notebaert E. Bench to
bedside review: iron metabolism in critically ill patients.
Crit Care, 2004; 8:356-62.
21. Garcia PC, Longhi F, Branco RG. Ferritin levels in
children with severe sepsis and septic shock, Acta
Paediatr, 2007; 96:1829-31.

89