April/May 2007

 

In this issue

Texas Children's continues its pioneering ways

Top med students match with Texas Children's

Ferritin levels and inflammatory diseases: What is significant for the Hemophagocytic Syndrome?

Creating a paperless trail, transforming health care

Grand Rounds

Medical staff committees and chairs

Home

Archives


Advisors

Ralph D. Feigin, M.D.
Physician-in-Chief
Texas Children's Hospital
Professor and Chairman
Department of Pediatrics
Baylor College of Medicine

Arnold G. Kagan, M.D.
Clinical Associate Professor of Pediatrics

ZoAnn E. Dreyer, M.D.
Medical Director,
Long-term Survivor Program
Texas Children's Hospital
Associate Professor
Baylor College of Medicine


Editor

Cindy Shanley
Marketing and Public Relations
Texas Children's Hospital
832-824-2180

 
 

For members of the Texas Children's Hospital medical staff

 

Ferritin levels and inflammatory diseases: What is significant for the Hemophagocytic Syndrome?

By Kenneth McClain M.D., Ph.D., Carl E. Allen, M.D., Ph.D., Xiaoying Yu, M.D., and Claudia A. Kozinetz, Ph.D., M.P.H.*

Note: This article is a summary of research done by Dr. Allen, a fellow in the Texas Children’s Cancer Center/Hematology Section

Ferritin is an iron-carrying protein used as a measure of body iron stores, even though the level present in the blood represents less than 1 percent of total body iron. Ferritin is also an acute phase reactant that is elevated in a number of inflammatory conditions including infections, shock, liver disease, peri-transplant mucositis, cystic fibrosis, as well as with iron overload from chronic transfusions. Markedly elevated ferritin levels have been associated with a rare but deadly hematologic disease: Hemophagocytic Lymphohistiocytosis (HLH).(1) In this condition, lymphocytes and macrophages are inappropriately active. Tissue biopsies show macrophages engulfing all types of blood cells in the bone marrow, lymph nodes, spleen, or liver.

HLH is caused by extemely high levels of cytokines (tumor necrosis factor alpha, interferon gamma, interlukin-1 alpha and others) that should normally mediate macrophage/lymphocyte interactions of a stable immune system. The primary molecular pathway for T cell regulation and NK cell granule processing have been disrupted by mutations in the perforin, MUNC or syntaxin genes in the familial or inherited forms of HLH. Secondary HLH, associated with any type of infection, immunizations, or unknown etiology, are sometimes found to have these mutations, but usually do not. How do you diagnose HLH? This is a syndrome for which the Histiocyte Society has designated eight diagnostic criteria of which at least five must be met:(2)

  1. Recurrent fevers

  2. Hepato/splenomegaly

  3. Cytopenias of at least two cell lines

  4. Hypofibrinoginemia OR hypertrygleridemia

  5. Hemophagocytosis in the bone marrow, lymph node or liver biopsy

  6. Defective NK cell function and decreased perforin expression (send out test)

  7. Elevated soluble IL-2 receptor (send out test)

  8. Ferritin over 500 µg/L

(Note by KM. Over 50 percent of these patients have CNS findings: lethargy, confusion, coma; CSF pleocytosis, elevated protein in CSF, MRI findings of meningeal enhancement or cerebral necrosis. When these are identified along with at least four of the above, I believe HLH should also be high on the differential diagnosis list.)

The diagnosis of HLH is not easy, since any one or a number of the diagnostic criteria may be found in other illnesses. Patients with HLH may be seen in our Emergency Department (ED) or in consult by the Infectious Disease, Intensive Care, Rheumatology or Hematolgy services which evaluate patients who present with fever of unknown origin, shock, hepatitis, Kawasaki Disease, or cytopenias. In 2003, Palazzi et al published a review of HLH patients seen at Texas Children’s Hospital by our Infectious Disease service and suggested an algorithm for evaluating patients with fever of unknown origin(3). HLH patients can progress from an ill-appearing child in the ED or regular ward to an extremely sick patient on a ventilator with pressor support and renal dialysis in a few hours. Thus, it is critical to identify patients with HLH and start the proper treatment (decadron, etoposide and cyclosporine) as quickly as possible. This is not to say that every child with a few of the above criteria has HLH because clearly shock, sepsis and hepatitis need to be fully evaluated and treated with supportive care no matter what the final diagnosis. The problem we often find is that hemophagocytosis is not clearly evident in a bone marow aspirate. In published studies, 20 percent of HLH patients will NOT have hemophagocytosis on the first marrow specimen, so clearly repeat marrow aspirates or biopsy of lymph node or liver is needed. Also, HLH is a disease that changes over time such that the cytokine stimulation of hemophagocytosis may be modest early on in the disease, or the marrow may become aplastic so no macrophages (histiocytes) are available to cause hemophagocytosis. Thus, finding hemophagocytosis is highly suggestive of HLH, but is neither necessary nor sufficient to make the diagnosis. HLH is a clinical syndrome that requires 5/8 diagnostic criteria be met before one can be sure that treatment should be started.

What can we do to improve our diagnostic acumen? Although no one criteria can make the diagnosis of HLH, a highly elevated serum ferritin can be very convincing, along with four other criteria. The Histiocyte Society has suggested a ferritin level over 500 µg/L because a survey found that most children with infectious diseases had levels less than this and those with rheumatologic diagnoses only rarely had higher levels. We were not satisfied with this criteria and decided to look at the diagnoses of all children with a ferritin level over 500 µg/L at Texas Children’s Hospital during a two year period (2003-2005). After receiving IRB approval for the study, we obtained the data and categorized children into 19 diagnostic groups:

Table 1.

1

HLH

2

Unknown

3

Shock

4

Stem cell transplant

5

Heart disease

6

Chronic transfusion

7

Immunodeficiency

8

Liver disease

9

Cystic fibrosis

10

Marrow failure

11

Prematurity

12

Malignancy

13

Viral infection

14

Bacterial infection

15

Solid organ transplant

16

Trauma

17

Autoimmune disease

18

Hemoglobinopathy

19

Renal disease

HLH patients had the highest ferritin levels, ranging from 994 to 189,721 µg/L with a mean of 44,984 µg/L. The next highest category was “Unknown,” where no specific diagnosis was made. Their ferritin levels ranged from 523 to 67,394 µg/L with a mean of 15,304 µg/L. The maximum values for shock (9066 µg/L), liver disease (12,937 µg/L), chronic transfusion (11,616 µg/L ) and autoimmune disease ( 37,407 µg/L) were all near or above the “magic” 10,000 µg/L level we had often associated with HLH in the past. However, the means and medians were all less than 6,000 µg/L. Statistical analysis showed that a ferritin over 10,000 µg/L was 90 percent sensitive and 96 percent specific for HLH. The addition of other HLH diagnostic criteria increased the specificity to 98 percent, but the sensitivity remained the same or decreased .

When we looked more closely at the “Unkown” category, there were three children with cardiopulmonary arrest of unknown cause (two died), four with fever of unknown origin – one of whom had a sibling die of HLH, one considered to have the macrophage activation syndrome, one suspected drug reaction to dilantin, and two for whom no chart could be located. None of the 10 children in the unknown group were fully investigated for HLH criteria. We suspect some of these children could have had HLH. HLH is likely underevaluated and underdiagnosed in children in general.

The serum ferritin is a convenient test since the results are available in our lab daily, whereas the IL-2 receptor and NK cell function test need to be sent to Cincinnati Children’s Hospital. When a child is suspected of having HLH, these two tests should be sent immediately, but treatment decisions may need to be made before this data is available, thus ferritin is a critical part of the diagnostic criteria.

During the two years of this study, there were 30,000 inpatient admissions at Texas Children’s Hospital, with 10 patients diagnosed with HLH. Thus, we believe that any tertiary pediatric hospital should expect to diagnose one HLH patient per 3,000 admissions annually. Over the past 20 years, our numbers have ranged from three to 11 per year.

The bottom line: If you see a patient with persistent fevers, hepato-splenomegaly and cytopenias of two cell lines, do the following: Obtain serum ferritin, AST/ALT, LDH, bilirubin, PT/PTT, fibrinogen and triglycerides. Consult the Hematology Service to perform a bone marrow aspirate and biopsy. If there are CNS symptoms, do a spinal tap. If four or more of the HLH diagnostic criteria are met, send labs for NK cell function and perforin staining of NK and T cells as well as the serum IL-2 receptor. Follow daily serum ferritin to see if the ferritin rises quickly (a strong indicator of HLH). Repeat bone marrow (or biopsy lymph node or liver) if first one negative for hemophagocytosis and clinical suspicion is high HLH. Rule out other conditions, though sepsis, viral infection, autoimmune disease and malignancy do not exclude a concurrent diagnosis of HLH. Initiate treatment for HLH sooner rather than later.

Kenneth McClain M.D., Ph.D., is a pediatric hematologist/oncologist with Texas Children’s Cancer Center® and professor of Pediatrics at Baylor College of Medicine. Carl E. Allen, M.D., Ph.D., a fellow in the Texas Children’s Cancer Center/Hematology Section. Xiaoying Yu, M.D., is a statistician in the Department of Pediatrics at Baylor College of Medicine. Claudia A. Kozinetz, Ph.D., M.P.H., is an associate professor of Pediatrics at Baylor College of Medicine.

References:

  1. McClain KL. Hemophagocytic Lymphohistiocytosis. Langerhans Cell Histiocytosis in Adults and Children in: UpToDate. B.D. Rose, ed. UpToDate Wellesley, MA. 2006.

  2. Henter JI, Horne A, Arico M, Egeler RM, Filipovich AH, Imashuku S, et al. HLH-2004: Diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer 2007;48:124-131.

  3. Palazzi DL, McClain KL, Kaplan SL. Hemophagocytic Syndrome in Children: An Important Diagnostic Consideration in Fever of Unknown Origin. Clinical Infectious Diseases 2003;36: 306-12.



Diagnostic Virology
Laboratory Newsletter

 

Back to top

Home     |     Contact us     |     Terms of use     |     Visit Texas Children's Hospital Web site     |     © 2006 Texas Children's Hospital