|Year : 2018 | Volume
| Issue : 2 | Page : 671-676
Evaluation of the role of adrenomedullin in febrile neutropenia
Fady M El-Gendy1, Mahmoud A El-Hawy1, Saly M El-Hefnawy2, Dina M. Adel Ahmed Shalaby3
1 Department of Pediatric, Faculty of Medicine, Menoufia University, Menoufia, Egypt
2 Department of Biochemistry, Faculty of Medicine, Menoufia University, Menoufia, Egypt
3 Department of Pediatrics, Benha Specialized Children Hospital, Benha, Egypt
|Date of Submission||01-Jan-2017|
|Date of Acceptance||27-Feb-2017|
|Date of Web Publication||27-Aug-2018|
Dina M. Adel Ahmed Shalaby
Shibin Al-Kom, Menoufia, 32511
Source of Support: None, Conflict of Interest: None
The aim of this study was to evaluate the role of adrenomedullin (ADM) as a new marker in febrile neutropenia (FN).
ADM has an effective role in FN episode in pediatric patients with acute lymphoblastic leukemia (ALL).
Patient and methods
This is a prospective study of 25 patients with ALL, recruited from the Hematology and Oncology Unit, Menoufia University Hospital, Egypt, from January to June 2016 diagnosed with ALL and suffering from FN episode. Complete blood count with differential count, C-reactive protein (CRP), ADM, Na, K, Ca, creatinine, and serum glutamate-pyruvate transaminase level were evaluated on the first and fifth days of FN.
ADM level was higher on the fifth day than on the first day of FN. It showed higher sensitivity and specificity compared with CRP and has greater area under the curve compared with CRP, with a cutoff of point 6 for CRP and 50 for ADM. CRP level was higher on the first than on the fifth day of FN.
ADM was higher on the fifth day than on first day of FN episode and has higher sensitivity and specificity compared with CRP and has greater area under the curve. Therefore, it can be considered as a new diagnostic marker in FN episode in patients with ALL. Nevertheless, CRP is still an effective cheap marker that should be used in FN episode.
Keywords: adrenomedullin, febrile neutropenia, pediatric patient with acute lymphoblastic leukemia
|How to cite this article:|
El-Gendy FM, El-Hawy MA, El-Hefnawy SM, Ahmed Shalaby DM. Evaluation of the role of adrenomedullin in febrile neutropenia. Menoufia Med J 2018;31:671-6
|How to cite this URL:|
El-Gendy FM, El-Hawy MA, El-Hefnawy SM, Ahmed Shalaby DM. Evaluation of the role of adrenomedullin in febrile neutropenia. Menoufia Med J [serial online] 2018 [cited 2018 Nov 20];31:671-6. Available from: http://www.mmj.eg.net/text.asp?2018/31/2/671/239730
| Introduction|| |
Leukemia is the most common type of cancer in children. All cancers begin in cells of the body, and leukemia is a cancer that begins in blood cell .
Leukemia is a cancer that involves the blood-forming tissues of the bone marrow, spleen, and lymph nodes. It is characterized by an uncontrolled production of abnormal, immature blood cells. There are different types of childhood leukemia. The most common type is called acute lymphoblastic leukemia (ALL) .
In this disease, too many underdeveloped infection-fighting white blood cells or lymphocytes are present in the blood and bone marrow of a child. ALL is the most common type of leukemia in children and the most common form of all types of cancer in children .
In childhood cancers, severe and prolonged neutropenia develops in association with the use of intensive chemotherapy protocols, which results in the onset of infections with potentially higher rates of morbidity and mortality .
Moreover, in childhood cancer, mortality rates associated with febrile neutropenia (FN) have decreased in recent years, but not to desired levels .
FN was defined as the presence of an axillary fever greater than 38.5°C once or greater than 38°C twice within 4 h in patients or with an absolute neutrophil count less than 500/mm 3 or between 500 and 1000/mm 3 that is expected to decrease down to less than 500/mm 3 within 24–48 h .
During the treatment of children with cancer, febrile reactions that can be observed after administration of chemotherapeutic agents, application of blood, and blood products can be confused with infection-related fever. Moreover, in bacterial and fungal infections with low virulence, the etiological agent cannot be identified in culture alone .
Therefore, clinical presentation and culture have a limited impact on the identification of the etiological agent of FN and on the intensity of needed treatment, precluding the application of unnecessary treatments .
C-reactive protein (CRP) and procalcitonin have been reported to be guiding tools for both the prediction of infection and the arrangement of treatment; thus, they are used prevalently .
Despite all of these biomarkers, severity of infection and risk for mortality cannot be fully determined. Therefore, new biomarkers have been investigated .
As a new biomarker, the predictive and prognostic values of adrenomedullin (ADM) in pulmonary, cardiovascular, and rheumatological diseases have been reported .
ADM is a vasoactive agent that is released in correlation with the severity of a systemic infection. It is a peptide with bactericidal activity and prognostic importance. Very few studies have examined its role in patients with FN .
The aim of this work was to evaluate the role ADM as a new marker in FN.
| Patients and Methods|| |
This is a prospective study of 25 patients with ALL recruited from the Hematology and Oncology Unit, Menoufia University Hospital, Egypt, from January to June 2016 diagnosed with ALL and suffering from FN episode.
Informed consent was obtained from all patients' guardians included in this study, which were approved by the local Ethical Committee. The study was approved by the Ethical Committee of Menoufia University.
Definitions used in our patients
FN: presence of an axillary fever greater than 38.5°C once or greater than 38°C twice and absolute neutrophil count of 500–1000/mm 3 that is expected to decrease down to less than 500/mm 3 within 24–48 h .
The inclusion criteria were as follows: both sexes, age younger than 18 years, diagnosis of malignancy, and febrile and neutropenic episode.
Any case not fulfilling the eligibility criteria was excluded from the study. All cases were subjected to a full clinical history taking and physical examination with special emphasis on the following.
- Personal history: name, age, and sex
- The complaint:
- Present history: fever and its onset, course and duration, abdominal pain with stress on its location, intensity, character, cough, appetite, nausea, vomiting, stool pattern, and consistency
- Past history: previous admission to the hospital and received treatment.
General examination included mental status, tachycardia, pallor, cyanosis, and jaundice.
Local examination included abdominal examination (for epigastria tenderness and loin tenderness), chest examination (for wheezes and crepitation), and neurological examination for convulsion.
Laboratory investigations included the following: (i) complete blood picture with differential count for the detection of neutropenia: mild neutropenia (1000: 1500 cells/mm 3), moderate (500: 1000 cells/mm 3,(severe (<500 cells/mm 3), or anemia ; (ii) CRP; (iii) blood culture for both bacteria and fungi; (iv) ADM serum level on days 1 and 5; (v) blood cultures based on clinical findings for both bacteria and fungi; and (vi) full liver function, kidney function, Ca, Na, and K.
Chest radiography was obtained.
Blood samples were drawn from patients on admission and day 5 to measure CRP, complete blood count (CBC), and ADM. In addition, 2 ml of venous blood was drawn from every child and then transferred into a plain tube and centrifuged for 10 min at 4000 rpm. The serum obtained was kept frozen at −20°C until analysis [measurement of serum ADM using enzyme-linked immunosorbent assay (ELISA)].
Acute phase reactant (CRP) was produced by hepatocytes in response to stimulation mediated by interleukins. CRP was estimated using the latex agglutination assay with the Rapi Tex CRP Commercial Kit (Omega Diagnostics Ltd,. Alloa, Scotland, UK) .
Blood culture for bacteria and fungi was prepared using a qualitative, automated (BACTEC 9050) (Becton Dickinson, Heidelberg, Germany) blood culture system on blood culture bottles for bacteria and fungi .
Human ADM is a vasoactive agent that is released in correlation with the severity of a systemic infection. It is a peptide with bactericidal activity and prognostic importance. Very few studies have examined its role in patients with FN . It was measured using an ELISA Kit (Sun Red, Germany) according to the manufacturer's recommendation .
Test principle was double-antibody sandwich ELISA to assay the level of human ADM in samples. ADM was added to monoclonal antibody enzyme well, which was precoated with human ADM monoclonal antibody and incubated. Thereafter, ADM antibodies labeled with biotin were added and combined with streptavidin–horseradish peroxidase to form an immune complex. Subsequently, incubation was carried out and the wells were washed again to remove the uncombined enzyme. Thereafter, chromogen solution A, B was added. The color of the liquid changes into blue and due to the effect of acid, the color finally becomes yellow. The chroma of color and the concentration of the human substance ADM in the sample are positively correlated .
The data were coded, entered, and processed on computer using SPSS (version 18, SPSS Inc, Chicago, Illinois, USA). The results were represented in tabular and diagrammatic forms and interpreted.
Mean, SD, range (minimum and maximum), and percentage were used as descriptive statistics.
The Spearman correlation test was used in the correlation between the parameters.
P value was considered significant as follows:
P value greater than 0.05: nonsignificant
P value less than 0.05: significant.
P value less than 0.001: highly significant.
All these tests were used as tests of significance at P value less than 0.05.
| Results|| |
This study included 25 cancer patients with an episode of FN recruited from the Hematology and Oncology Unit at Menoufia University Hospital, Egypt, from January to June 2016 diagnosed with ALL and suffering from FN episode. Their ages ranged from 1 to 14 years with a mean ± SD of 4.79 ± 3.81 year. Sex distribution was as follows: 15 (60.0%) male and 10 (40.0%(female patients; 14 (56%) were from rural area and 11 (44%) from urban area [Table 1].
As regards the organisms causing infection in FN attack, 17 (68%) were Gram negative bacilli and 8 (32%) were fungal infection.
All cases presented with chest infection; 14 (56%) radiographs showed pneumonia, whereas the other 11 (44%) did not show pneumonia.
ADM was significantly higher on the fifth day than on the first day of FN (P = 0.000), but CRP was significantly higher on the first day than on the fifth day of FN (P = 0.021) [Table 2].
|Table 2: Adrenomedullin and C-reactive protein on the first and the fifth day of febrile neutropenia|
Click here to view
There was no significant correlation between CRP and days of neutropenia or type of culture [Table 3]. There was no significant correlation between CRP and ADM on the first or the fifth day of FN [Table 3].
|Table 3: Correlation between C-reactive protein and days of neutropenia and type of culture and also correlation between C-reactive protein and adrenomedullin on the first and the fifth day of febrile neutropenia|
Click here to view
White blood cells, platelets, and neutrophils were significantly higher on the fifth day than on the first day, whereas there was no significant difference in hemoglobin between the first and the fifth day [Table 4]. The level of Na, K, creatinine, and serum glutamate-pyruvate transaminase (SGPT) showed no significant changes on the first and the fifth day of FN.
|Table 4: Complete blood count of the selected group on the first and the fifth day of febrile neutropenia|
Click here to view
ADM had a higher sensitivity (93.33%) compared with CRP (80%) in FN episode. As regards the specificity of the markers in FN, ADM was more specific (85.4%) compared with CRP (73%). The receiver operating characteristics (ROC) curve also showed that the positive predictive value of ADM in this study was 90%, which was higher compared with the positive predictive value of CRP (89%). The area under the curve (AUC) in ADM was 0.94, whereas in CRP it was 0.76 [Table 5].
|Table 5: Comparison between the sensitivity and specificity and predictive value of adrenomedullin and C-reactive protein level in febrile neutropenia|
Click here to view
There was no significant correlation between CRP and Na, K, creatinine, and SGPT on the first day of FN. Moreover, there was no significant correlation between CRP and CBC parameters on the first day of FN. There was no significant correlation between CRP and Na, K, creatinine, and SGPT on the fifth day of FN. Moreover, there was no significant correlation between CRP and CBC parameters on the 5th day of FN.
There was no significant correlation between ADM and Na, K, creatinine, and SGPT on the first day of FN. Moreover, there was no significant correlation between ADM and CBC parameters on the first day of FN.
There was no significant correlation between ADM and Na, K, creatinine, and SGPT on the fifth day of FN. Moreover, there was no significant correlation between ADM and CBC parameters on the fifth day of FN.
| Discussion|| |
In the present study, the age of patients ranged between 1 and 14 years with a mean ± SD of 4.79 ± 3.81 and a median of 5 years. As regards sex, 60% of patients were male and 40% were female. Similarly, in the study by El Rashedy et al.  the age in their study ranged from 1 to 14 years with a mean ± SD age of 6.4 ± 4.0 years.
Almost all cases during the episode of FN presented with chest manifestation. This is in agreement with the findings of Park and Kim , who mentioned that respiratory tract infection was the most common manifestation .
In this study, 56% of chest radiographs showed abnormalities (pneumonia). This is in agreement with the findings of Moon and Chun , who reported that pulmonary infiltration on chest radiography at presentation could be used to identify FN patients who will develop complications, and this factor may be useful in making treatment-related decisions in the emergency department.
Total leukocytic count in this study showed a significant increase on the fifth day than on the first day of FN episode, but there was no correlation between it and CRP. This is in agreement with the findings of El-Mahallawy et al. , who mentioned that there was no relation between reduced total leukocytic count and the infection in children with cancer. In contrast, Ammann et al.  reported that total leukocytic count is a risk factor for predicting adverse events in pediatric oncology patients with FN.
The absolute neutrophilic count in this study was lower on the first day than on the fifth day. Moreover, there was no correlation between it and CRP. This is in agreement with the findings of Póvoa et al. , who mentioned that the CRP course seems to be independent from the presence or absence of neutropenia.
Platelets showed a significant difference in this study between the first and the fifth day, but no correlation between it and CRP. In contrast, Miyamoto et al.  mentioned that platelets had a significant correlation between it and CRP in cancer progression.
As regards hemoglobin level, there was no difference between the first and the fifth day. Moreover, there was no correlation between it and CRP. In contrast, Aydogdu et al.  reported that there was a significant correlation between hemoglobin and CRP.
In this study, the incidence of Gram-negative organisms was 68%. This is in agreement with the study by Kanafani et al. , who showed that gram-negative bacteria remain the predominant pathogens in multiple centers.
In this study, eight attacks of FN out of 25 (32%) attacks had positive fungal blood culture. This is in agreement with the findings of El-Beblawy and Awad , who found that the rate of fungal colonization and infection in infants and children with hematological malignancies was high.
CRP in this study was significantly higher on the first day than on the fifth day of FN. This is in agreement with the findings of Sugiura et al. , who found that CRP showed elevation in FN episode.
ADM level in this study was significantly higher on the fifth day than on the first day, and also showed higher sensitivity and specificity (93.3 and 85.4%) compared with CRP (80 and 73%) [Table 5], It had an AUC of 0.94, which was greater than that for CRP (0.76). The cutoff point was 6 for CRP and 50 for ADM, as illustrated in ROC diagram for ADM and CRP in febrile neutropenia.
This is in agreement with the findings of Demirkaya et al. , who detected significantly higher levels of ADM on day 7 than on third and first days of FN, and they also found that the sensitivity and specificity and accuracy of ADM was higher compared with CRP in the diagnosis and follow-up of FN episode.
Moreover, this is in agreement with the findings of Debiane et al. , who found that ADM has a promising role in predicting blood stream infections in a manner more helpful compared with CRP. This biomarker was superior to CRP in the prognostic analysis in febrile patients with cancer and showed increasing levels on day 4 than on day 1. Moreover, it had a greater AUC ROC (0.70) compared with CRP (0.53) with a cutoff point of 6 for CRP and 60 for ADM .
Moreover, Van der Starre et al.  showed that the diagnostic accuracy for predicting the severity of infection was reflected by the AUC ROC and was higher in ADM (0.83), whereas CRP (0.56) and other markers lacked diagnostic value in this respect. They reported a cutoff point of 8 for CRP and 50 for ADM .
Moreover, Enguix et al.  found that ADM showed a better prognostic value and had higher accuracy (90%) in febrile episode in comparison with CRP (77%). They also showed a cutoff point of 8 for CRP and 66 for ADM .
It is also in agreement with the findings of Nishida et al. , who reported that plasma ADM is a powerful independent predictor for future events in high-risk patients, suggesting that its predictive value is superior to that of CRP. They also showed a cutoff point of 6 for CRP and 55 for ADM .
This is in agreement with the findings of Di Liddo et al. , who found that ADM level increased in malignancy and ADM also showed high sensitivity and specificity in cancer diagnosis and progression.
In contrast, Kesik et al.  found that the changes in ADM levels through time periods ( first, second, third, and seventh days of FN) were not significant. This may be due to the small number of patients in their study .
It is in disagreement with the findings of Dötsch et al. , who found that plasma ADM level increased nonspecifically during illnesses and is not necessarily associated with sepsis in critically ill and malignant patient and is linked mostly to tumor differentiation but not to prognostic markers .
In this study, the duration of neutropenia had no significant correlation between it and CRP or type of infection. This is in agreement with the findings of Park and Kim , who reported that the duration of neutropenia showed no significant relationship with serious infectious complications.
There was no correlation between CRP and ADM. This is in agreement with the findings of Demirkaya et al. , who found no correlation between CRP and ADM.
| Summary and Conclusion|| |
This study was conducted to evaluate the role of ADM in FN episode in pediatric patients with ALL, and this study showed that ADM was higher on the fifth day than on the first day, and hence it can be considered as a new diagnostic marker in FN episode in patients with ALL. Nevertheless, CRP is still an effective cheap marker that should be used in FN episode.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Karakurt DG, Demirsoy U, Corapcioglu F. Pro inflammatory cytokine levels predict serious complication risk of infection in pediatric cancer patients. Pediatr Hematol Oncol 2014; 31
Phillips RS, Wade R, Lehrnbecher T. Systematic review and meta-analysis of the value of initial biomarkers in predicting adverse outcome in febrile neutropenic episodes in children and young people with cancer. BMC Med 2012; 10
Semeraro M, Homée C, Rolland E. A predictor of unfavorable outcome in neutropenic paediatric patients presenting with fever of unknown origin. Pediatr Blood Cancer 2010; 54
Haeusler GM, Carlesse F, Phillips RS. An updated systematic review and meta-analysis of the predictive value of serum biomarkers in the assessment of fever during neutropenia in children with cancer. Pediatr Infect Dis J 2013; 32e
Santolaya ME, Alvarez AM, Aviles CL. Admission clinical and laboratory factors associated with death in children with cancer during a febrile neutropenic episode. Pediatr Infect Dis J 2007; 26
Paganini HR, Aguirre C, Puppa G. A prospective multi centric scoring system to predict mortality in febrile neutropenic children with cancer. Cancer 2007; 109
Lehrnbecher T, Phillips R, Alexander S. International Pediatric Fever and Neutropenia Guide-line Panel. Guideline for the management of fever and neutropenia in children with cancer and/or undergoing hematopoietic stem-cell transplantation. J Clin Oncol 2010; 30
Kim DY, Lee YS, Ahn S. The usefulness of procalcitonin and C-reactive protein as early di-agnostic markers of bacteremia in cancer patients with febrile neutropenia. Cancer Res Treat 2011; 43
Mian A, Becton D, Saylors R. Biomarkers for risk stratification of febrile neutropenia among children with malignancy: a pilot study. Pediatr Blood Cancer 2012; 59
Schuetz P, Christ-Crain M, Morgenthaler NG. Circulating precursor levels of endothelin-1 and adrenomedullin, two endothelium-derived, counteracting substances, in sepsis. Endothelium 2007; 14
Al Shuaibi M, Bahu RR, Chaftari AM. Pro-adrenomedullin as a novel biomarker for predicting infections and response to antimicrobials in febrile patients with hematologic malignancies. Clin Infect Dis 2013; 56
Fanci R, Paci C, Martinez RL, Fabbri A, Pecile P, Leoni F, Longo G. Management of fever in neutropenic patients with acute leukemia: current role of ceftazidime plus amikacin as empiric therapy. J Chemother 2000; 3
Park Y, Kim DS. The suggestion of a risk stratification system for febrile neutropenia in patients with hematologic disease. Leuk Res 2010; 34
El Rashedy F, Ragab S, Dawood A. Variation in efficacy of 6 mercaptopurine in pediatric patient with acute lymphoblastic leukemia. MMJ 2014; 6
Moon JM, Chun BJ. Predicting the complicated neutropenic fever in the emergency department. Emerg Med J 2009; 26
El-Mahallawy H, Mohamed NH, Attia EA. Risk factor evaluation and prognosis of bacteremia in children with cancer. J Egypt Natl Canc Inst 1999:4
Ammann RA, Bodmer N, Hirt A. Predicting adverse events in children with fever and chemotherapy-induced neutropenia: The prospective multicenter SPOG 2003 FN study. J Clin Oncol 2010; 28
Póvoa P, Souza-Dantas VC, Soares M. C-reactive protein in critically ill cancer patients with sepsis: influence of neutropenia, Crit Care 2011: 15
Miyamoto R, Oda T, Hashimoto S. Platelet × CRP multiplier value as an indicator of poor prognosis in patients with resect able pancreatic cancer, J Neuroendocrine Tumors Pancreat Sci 2017; 46:
Aydogdu O, Bozkurt IH, Yonguc T. Re: impact of preoperative hemoglobin and CRP levels on cancer-specific survival in patients undergoing radical cystectomy for transitional cell carcinoma of the bladder: results of a single-center study. World J Urol 2017; 35
Kanafani ZA, Dakdouki GK, El-Chammas KI. Bloodstream infections in febrile neutropenic patients at a tertiary care center in Lebanon: a view of the past decade. Int J Infect Dis 2007; 11
El-Beblawy N, Awad AS. Fungal colonization and infection in childhood acute leukemia and lymphoma during induction therapy. Egypt J Pediatr 2001; 8
Sugiura S, Ito T, Koyama N. Asymptomatic C-reactive protein elevation in neutropenic children, Pediatr Int 2017; 59
Demirkaya M, Tugcu D, Akcay A. Adrenomedullin – a new marker in febrile neutropenia: comparison with CRP and procalcitonin. Pediatr Hematol Oncol 2015; 32:482–489.
Debiane L, Hachem RY, Al Wohoush I. The utility of adrenomedullin and procalcitonin in comparison to C-reactive protein as predictors of sepsis and bloodstream infections in critically ill patients with cancer. Crit Care Med 2014; 12
Van der Starre WE, Zunder SM, Vollaard AM. Prognostic value of adrenomedullin, procalcitonin and C-reactive protein in predicting outcome of febrile urinary tract infection. Clin Microbiol Infect 2014; 20
Enguix A, Escobar R, García DA. Usefulness of several biomarkers in the management of septic patients: C-reactive protein, procalcitonin, presepsin and adrenomedullin. Clin Chem Lab Med 2016; 54
Nishida H, Horio T, Suzuki Y. Plasma adrenomedullin as an independent predictor for future high-risk patients: comparison with C-reactive protein and adiponectin. Peptides 2008; 29
Di Liddo R, Bridi D, Gottardi M. Adrenomedullin in the growth modulation and differentiation of acute myeloid leukemia cells. Int J Oncol 2015; 48
Kesik V, Ataş E, Gülcan Kurt Y. Adrenomedullin predicts high risk and culture positivity in children with solid tumors suffering from neutropenic fever. J Infect Chemother 2016; 9
Dötsch J, Harmjanz A, Christiansen H. Gene expression of neuronal nitric oxide synthase and adrenomedullin in human neuroblastoma using real-time PCR. Int J Cancer 2000; 88:
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]