Feature Articles
Therapeutic Role of Anakinra, an Interleukin-1 Receptor Antagonist, in the Management of Secondary Hemophagocytic Lymphohistiocytosis/ Sepsis/Multiple Organ Dysfunction/Macrophage Activating Syndrome in Critically Ill Children* Surender Rajasekaran, MD, MPH1; Katherine Kruse, MD1,2; Karen Kovey, PharmD1; Alan T. Davis, PhD2; Nabil E. Hassan, MD1; Akunne N. Ndika, MBBS, MPH1; Sandra Zuiderveen, BSN, RN1; James Birmingham, MD3
Objectives: Secondary hemophagocytic lymphohistiocytosis, macrophage activating syndrome, and sepsis share the same inflammatory phenotype leading often to multiple organ dysfunction syndrome needing intensive care. The goal of this article is to describe our experience with anakinra (Kineret), a recombinant interleukin-1 receptor antagonist, in decreasing the systemic inflammation. Design: Retrospective case series. Setting: The PICU at the Helen DeVos Children’s Hospital (Grand Rapids, MI). Patients: The records of eight critically ill children presumed to have secondary hemophagocytic lymphohistiocytosis at our institution between January 1, 2011, and July 31, 2012, were reviewed. Interventions: All of the patients were treated with anakinra (Kineret) and in some cases systemic corticosteroids as first-line therapy for secondary hemophagocytic lymphohistiocytosis. Measurements and Main Results: Patients had a median age of 14 years and a median Pediatric Risk of Mortality score of 11.5. Four were previously healthy and four had underlying diseases that could have made them susceptible to secondary hemophagocytic lymphohistiocytosis. Indications for PICU transfer were respiratory distress 50% (4 *See also p. 486. 1 Pediatric Critical Care Medicine, Helen DeVos Children's Hospital. Grand Rapids, MI. 2 Grand Rapids Medical Education Partners, Grand Rapids, MI. 3 Department of Rheumatology, Helen DeVos Children's Hospital. Grand Rapids, MI. The authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: surender.rajasekaran@ helendevoschildrens.org Copyright © 2014 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies DOI: 10.1097/PCC.0000000000000078
Pediatric Critical Care Medicine
of 8), cardiovascular instability 37.5% (3 of 8), and chest pain (1 of 8). Five of the patients (62.5%) were mechanically ventilated and 62.5% (5 of 8) received vasoactive infusions. Inflammatory markers were assessed linearly at the start of therapy and 7 days later. Baseline C-reactive protein was 206 ± 50 mg/L (mean ± sem) at the start of anakinra and decreased by 67.1% to 68 ± 36 mg/L (p = 0.03). Ferritin decreased by 63.8% to 3,210 ± 1,178 ng/mL (p = 0.30), and fibrinogen decreased by 42% to 158 ± 41 mg/dL (p = 0.03). Absolute neutrophil count (p = 0.38) and absolute lymphocyte count (p = 0.69) did not change significantly. No infections were attributed to anakinra therapy. One patient died long after treatment with anakinra while receiving pre-hematopoietic stem cell transplant chemotherapy. Conclusions: Anakinra could represent a promising therapeutic approach in these life-threatening disorders that are likely underdiagnosed and often difficult to treat. (Pediatr Crit Care Med 2014; 15:401–408) Key Words: cytokines; hemophagocytic lymphohistiocytosis; inflammation; macrophage activating syndrome; multiple organ dysfunction syndrome; pharmacotherapy
F
or the last 2 decades, the unstable patient with hemophagocytic lymphohistiocytosis (HLH) has been treated using systemic corticosteroids and etoposide in most centers. This regimen, developed in 1994 by the histiocyte society through expert consensus, continues to be used despite its many known side effects (1). The society has broadened its diagnostic criteria in order to make an earlier diagnosis and it is now recognized that significant clinical and laboratory overlap occurs between secondary HLH/macrophage activating syndrome (MAS) and sepsis/systemic inflammatory response syndrome (SIRS)/multiple organ dysfunction syndrome www.pccmjournal.org
401
Rajasekaran et al
(MODS) (2–5). The risk of applying any HLH/MAS-targeted therapies to children with sepsis/SIRS/MODS has to be considered whatever therapeutic modality is eventually chosen, as patients with sepsis/SIRS might suffer harm from a therapy that targets inflammation (3). Therefore, it is critical that the risks of treatment or nontreatment be weighed according to the clinical presentation of each patient. As such it would be helpful if the therapy is tolerated by all the grouped disorders. Patients with HLH spectrum of disease often present with nonspecific signs and symptoms that mimic viral or bacterial illness. For the intensivist, it is important to understand the diagnostic criteria and management strategies because the vast majority of these HLH diagnoses are made in the PICU (6). Early diagnosis remains challenging and is often subjective and should be made in consultation with other subspecialties. The primary diagnosis is based on a series of predetermined clinical and laboratory criteria derived from the HLH-2004 consensus (7). It is recommended that initial therapy be the same regardless of HLH type (8). Nevertheless, an understanding of the various types of HLH may assist in developing management strategies and understanding prognosis. The Subtypes of HLH Patients in the “primary” HLH category are those with familial inheritance or genetic causes, are usually younger, and are
thought to have fixed defects of cytotoxic cell function. They have a clear risk of HLH recurrence and are not likely to survive without hematopoietic stem cell transplantation (HSCT). Although HLH in these patients can also be associated with infections, the immunologic trigger may not be apparent. The term “secondary HLH” or “reactive HLH” generally refers to older children who present without a family history or known genetic cause for their HLH. These patients typically have concurrent infections or chronic illness that appears to trigger the HLH (8). MAS is a term typically used to describe HLH associated with rheumatologic disorders but is likely not pathophysiologically distinct from secondary or reactive forms of HLH. For the purposes of our discussion, we will use the term “secondary HLH” to encompass the spectrum disorders under our consideration for early identification and management. Therapy is directed at ameliorating inflammation by attenuating the hypercytokinemia (or what has been termed the “cytokine storm”) while simultaneously supporting failing organs. Steroids, chemotherapy, and stem cell transplantation are treatment options, but these often lead to opportunistic infections and further organ damage. The clinical and laboratory findings in HLH can be anticipated based on the known biological effects of the inflammatory cytokines tumor necrosis factor-α and interleukin (IL)-1. Some studies have shown elevated serum IL-1 in patients with HLH (9). As would be
Table 1. Patient Characteristics and Laboratory Values Performed Within 72 Hours of Anakinra Initiation Patient
1
Age (yr)/Sex
15/M
Preexisting Conditions
Hyper IgM CD40 ligand deficiency
Organ/Lymph Node
Bone Marrow/Tissue Phagocytes
Natural Killer Activity
Ferritin Prior (ng/mL)
Liver
None
Not done
19,790
Spleen LN 2
20/M
Healthy
None
+
Low
23,320
3
13/M
Healthy
Liver
++
Low
1,138
Lymph node
Normal
623
Spleen
No bone marrow done
Low
683
LN
Peripheral cytometry
LN 4
15/F
Family history of systemic lupus erythematosus
Liver Spleen LN
5
21/F
Renal transplant 2009 lupus variant
6
11/M
Healthy
None
++
Low
28,931
7
8/M
Liver transplant acetyl CoA deficiency
Liver
+
Not done
1,372
+
Not done
6,651
Spleen 8
9/F
Healthy
Liver Spleen LN
LN = lymphadenopathy, + = few hemophagocytic histiocytes, ++ = frequent hemophagocytic histiocytes. a Triglyceride the highest from the first 72 hr of therapy.
402
www.pccmjournal.org
June 2014 • Volume 15 • Number 5
Feature Articles
hypothesized, several case reports and small series have demonstrated benefits of IL-1 antagonist therapy in patients with secondary HLH. However, these have been either single patient reports or one small series involving patients with juvenile idiopathic arthritis (JIA) that developed MAS (10, 11). In this article, we examine the clinical course of eight critically ill pediatric patients with presumed or confirmed secondary HLH who were treated with anakinra.
METHODS Following approval by the institutional review board, we performed a retrospective review of eight patients with suspected secondary HLH who received care in our PICU at Helen DeVos Children’s Hospital between January 2011 and July 2012. These patients were ultimately treated with anakinra as part of their management. A retrospective review of the Cerner database was conducted to identify all patients who were worked up for or identified as having secondary HLH. Once identified, their clinical course was plotted prior to and after initiation of anakinra. Virtual PICU systems and Cerner were used to determine the clinical course and laboratory indices. Patients included in our analysis were PICU patients diagnosed with secondary HLH and received anakinra as part of the therapy. Patients with secondary HLH not requiring PICU admission were excluded from our review. Various clinical and
Ferritin Highest (ng/mL)
Cytopenia > 2 Cell Lines
Elevated Liver Enzymes
No
Yes
57
287
12,000
No
No
No
114
174
1,207
No
4,757 (day 4)
No
Yes
238
369
20,000
No
1,552 (day 3)
Yes
Yes
186
258
3,066
No
1,100 (day 3)
Yes
Yes
134
322a
11,128
No
No
No
44
343
2,213
No
Yes
Yes
76
168a
12,220
Yes
No
Yes
415
25,360 (day 3)
Pediatric Critical Care Medicine
Fibrinogen (mg/dL)
laboratory findings were extracted from the specified databases and presented in the Results section. Typically, the patients with suspected HLH are managed in consultation with infectious disease, rheumatology, and hematology providers. The diagnosis was ultimately decided in conjunction with evaluation by the pediatric rheumatologist who then determined when to initiate anakinra. Numerous factors were considered in the determination of diagnosis and the timing of starting therapy with anakinra; however, several factors were universal including fever, laboratory and clinical evidence of uncontrolled inflammation (hyperferritinemia was an absolute requirement), lack of a definitive infectious source, and evidence of clinical deterioration. The diagnosis of secondary HLH was made with consideration of the criteria established by the HLH-2004 consensus (7) that identified the following criteria: familial history or known genetic defect and clinical/laboratory criteria including fever, splenomegaly, cytopenia ≥ 2 cell lines, hemoglobin < 9 g/dL, platelets < 100,000/L, neutropenia < 1 × 109/L, triglyceridemia, ferritin ≥ 500 ng/mL, fibrinogen < 150 mg/ dL, soluble CD25, decreased natural killer (NK) cell activity, hemophagocytosis in bone marrow (BM), lymph node, and cerebrospinal fluid (CSF), and elevated transaminases, bilirubin, lactate dehydrogenase (LDH), CSF, pleocytosis, or protein (7). Organ system dysfunction was defined according to the criteria established by Wilkinson et al (12) and subsequently
Triglyceride (mg/dL)
246
Lactate Dehydrogenase (IU/L)
Cerebrospinal Fluid Pleocytosis
No
www.ccmjournal.org
403
Rajasekaran et al
modified by Proulx et al (13). BM or other tissue confirmation was attempted in all patients, but incomplete or partial fulfillment of the criteria was deemed sufficient for therapeutic trial of anakinra when an alternate diagnostic explanation for the individual’s clinical status was not identified. Dosing of Anakinra Anakinra (Kineret) is available in single-use 1-mL prefilled glass syringes for daily subcutaneous (SC) administration. Each 1-mL prefilled glass syringe contained 0.67 mL (100 mg) of anakinra in a solution (pH 6.5) containing sodium citrate (1.29 mg), sodium chloride (5.48 mg), disodium EDTA (0.12 mg), and polysorbate 80 (0.70 mg) in water for injection, USP. Dosing in our patients was determined by the prescribing rheumatologist. The dose was initially based on patient weight but was rapidly increased based on evaluation of the clinical response. It should be noted that the usual recommended dose of anakinra (usually between 1 and 2 mg/kg/d) is for patients with JIA and is not necessarily applicable in this setting. We were comfortable using a higher than commonly recommended doses of anakinra for several reasons. First, the half-life of anakinra SC is very short (4–6 hr), justifying more frequent than daily dosing in the setting of acute systemic inflammation, in order to rapidly achieve therapeutic IL-1 inhibition. Additionally, prior experience with systemic juvenile arthritis patients had shown that children often require higher doses, as much as 11 mg/kg/d (14). Finally, there is published experience using high doses of continuous IV IL-1 receptor antagonist (IL-1ra) in sepsis patients, providing support for the safety of higher doses in an SIRS setting, even without evidence of
Figure 1. Clinical course of the patients.
404
www.pccmjournal.org
any compelling therapeutic effect in that analysis (15). Dosing was maintained until serological markers of inflammation were noted to be declining and this correlated with clinical improvement, further adjustments in the dosing of anakinra were determined by the rheumatologist. In addition to above therapy, five patients received IV immune globulin (IVIG) therapy and six patients received high-dose corticosteroid therapy with doses ranging from 10 to 20 mg/kg/d. Kineret prescribing information is available at http:// www.accessdata.fda.gov/drugsatfda_docs/label/2003/anakamg062703LB.pdf. Statistical Methods Summary statistics were calculated. Quantitative data are expressed as the mean ± sem, whereas nominal data are expressed as a percentage. For various laboratory indices, comparisons between the time period just prior to administration of anakinra (anakinra start) and 1 week postadministration were performed using the Wilcoxon signed rank test. Significance was assessed at p less than 0.05. Analyses were performed using SPSS v.21 (IBM, Armonk, NY).
RESULTS Eight patients admitted to the PICU for HLH from January 2011 to July 2012 were included for analysis. Five of the eight patients were mechanically ventilated for respiratory failure and five patients needed vasoactive infusions to maintain cardiovascular function. One patient needed renal replacement therapy (RRT) and another with the most prolonged course died 217 days post PICU admission. Patient Characteristics Patient characteristics, including most of the clinical and laboratory features used to make the initial diagnosis of secondary HLH, are described in Table 1. Eight patients (five males and three females) with an average age of 13.8 ± 1.6 years were diagnosed with secondary HLH. Four patients had preexisting diseases and four were previously healthy. Six patients had bone marrow aspiration (BMA) done with only two demonstrating significant hemophagocytosis and three showing some phagocytosis. Patient 4 did have a lymph node examination during the admission which was June 2014 • Volume 15 • Number 5
Feature Articles
Diagnostic Findings of Patients With Hyperferritinemia and Secondary Hemophagocytic Lymphohistiocytosis/ Sepsis/Multiple Organ Dysfunction Syndrome/Macrophage Activating Syndrome
making the initial assessment of secondary HLH (Table 1). Later, ferritin and CRP levels were used to evaluate therapy. Six of the patients received steroids with anakinra and five received IVIG. The dosing and duration of anakinra were determined as noted earlier. Anakinra was started at median day 5 of hospitalization (range, 3–13) and median duration of anakinra administration was 8.5 days (range, 5–31 d; Fig. 1). The single mortality was ultimately suspected of having primary HLH (no perforin 1 mutations detected) or at least a more aggressive form of HLH. Therapy was changed to the typical HLH protocol including etoposide within 2 weeks of discontinuing anakinra as his symptoms quickly recurred. By the time of death, he had been off anakinra for months (Fig. 1).
Table 2.
Secondary Hemophagocytic Lymphohistio cytosis/Macrophage Activating Syn drome (%) (n = 8)
Variable
Fever
100 (8)
Lactate dehydrogenase > 1,000 IU/L
100 (8)
Hyperferritinemia > 500 ng/mL
100 (8)
Bicytopenia
87 (7)
Anemia < 9 g/dL
87 (7)
Survival
87 (7)
Thrombocytopenia platelet < 100/μL
62 (5)
Hypertriglyceridemia > 265 mg/dL
62 (5)
Hemophagocytosis in bone marrow
62 (5)
Splenomegaly
50 (4)
Hypofibrinogenemia < 150 mg/dL
50 (4)
Neutropenia absolute neutrophil count < 1,000/μL
25 (2)
Markers of Inflammation and Other Laboratory Indices The clinical markers suggestive of secondary HLH were found in varying degrees with fever, elevated ferritin (> 500 ng/mL) and elevated LDH (> 1,000 IU/L) being found in 100% of patients (Table 2). The concentrations of a variety of markers of inflammation at various time points before and after anakinra administration are described in Table 3. All seven of the markers listed showed declines in their values from just prior to administration of anakinra (anakinra start) to 1 week after administration, although this change was only statistically significant for CRP and fibrinogen (p = 0.03). Mean serum ferritin decreased by 63.8% during this time period (p = 0.30), absolute neutrophil count (ANC) decreased by 30.2% (p = 0.38), and platelets by 29.3% (p > 0.99).
All of the patients met five or more of the criteria required for a diagnosis of hemophagocytic lymphohistiocytosis/macrophage activating syndrome.
not typical for HLH. Patient 5 did not receive a BMA as the attending on service felt the procedure could compromise the vascular access that was necessary for RRT in this fluid overloaded patient. Five patients had peripheral NK cells examined for perforin and granzyme B done for clinical reasons. The levels of serum triglyceride, ferritin, and C-reactive protein (CRP) in the first 72 hours of therapy were noted, as they assisted in Table 3.
Patients’ ICU Characteristics The patients had a median Pediatric Risk of Mortality score of 11.5 and spent an average of 9.3 ± 1.7 days in the PICU (Table 4). Patients 3, 4, and 8 were directly admitted into the ICU from out-of-hospital referring centers. Five patients had multiple organ involvement 48 hours into ICU admission necessitating vasoactive infusions, five needed ventilatory support, and one required RRT.
Laboratory Markers in Patients Treated With Anakinra
Variables
Admission
C-reactive protein (mg/L)
150 ± 28
72 Hr Post
1 Wk
2 Wk
206 ± 50a
98 ± 37
68 ± 36a
30 ± 9
8,098 ± 4,696
9,761 ± 4,212
7,300 ± 3,134
3,210 ± 1,178
1,464 ± 425
Fibrinogen (mg/dL)
508 ± 105
271 ± 108b
119 ± 32
158 ± 41b
267 ± 86
Creatinine (mg/dL)
0.7 ± 0.2
0.9 ± 0.3
1.0 ± 0.3
0.7 ± 0.1
0.6 ± 0.2
Lactate dehydrogenase (IU/L)
17,747 ± 17,045
2,593 ± 1,552
1,458 ± 690
994 ± 463
271 ± 59
Absolute neutrophil count (103/μL)
8,843 ± 5,550
6,854 ± 2,384
7,407 ± 2,356
4,786 ± 1,724
1,023 ± 494
229 ± 111
289 ± 103
327 ± 116
204 ± 60
235 ± 58
Ferritin (ng/mL)
Platelets (103/μL)
Anakinra Start
C-reactive protein showed statistically significant declines at 1 wk, in comparison to the values at anakinra start (p = 0.03). b Fibrinogen showed statistically significant declines at 1 wk, in comparison to the values at anakinra start (p = 0.03). None of the other variables showed significant changes for the same time interval (p > 0.05). Only three patients had complete blood counts at 2 wk. Values reported as the mean ± sem. a
Pediatric Critical Care Medicine
www.ccmjournal.org
405
Rajasekaran et al
The vasoactive infusions used to maintain perfusion included three patients receiving norepinephrine and milrinone, one receiving only milrinone, and one receiving dopamine. Mechanical ventilation was for a median of 6 days (range, 2–11 d; Fig. 1), and one patient (patient 3) received 2 L/min nasal cannula oxygen. All eight patients had chest radiograph changes (Table 4). The starting mean oxygenation index 12 hours after intubation and initiation of ventilation was 15.2. Initiation of anakinra was varied but the median start time was day 1 of mechanical ventilation (Fig. 1). One patient with a history of previous renal transplant needed to be on continuous RRT during her ICU course. Recovery in relation to anakinra administration is illustrated in Figure 1. Anakinra and Infection Patients 1, 2, 5, and 7 had infections around their anakinra course (Table 4). These infections were also possibly the triggers for the secondary HLH. Patient 6 had two infections: the first was 9 days after anakinra was stopped and the next one was prior to mortality. Patients were administered anakinra along with specific antimicrobials wherever indicated. Patient 1 had a histoplasma infection that was thought to be the trigger for secondary HLH, even though Figure 1 shows that the diagnosis occurred days into the anakinra course. At diagnosis, a single patient was neutropenic with 450 cells/μL. Later at 1 week, three patients became neutropenic (WBC < 1,500 cells/μL) and one patient became lymphopenic (absolute lymphocyte count [ALC] < 1,000 cells/μL) at day 7 of anakinra. Two of the three neutropenic patients were also on systemic steroids. The one patient who died had normal ANC and ALC during and immediately after his course of anakinra.
DISCUSSION Patients with secondary HLH present with the same phenotypical features as sepsis. Distinguishing them is important because the management of sepsis is one of intensive support Table 4.
while providing the appropriate antibiotic coverage whereas that for secondary HLH is to ameliorate inflammation. Considering these pathophysiologically related disorders together as sepsis/SIRS/MODS/MAS until a more specific diagnosis is made would encourage that the diagnosis of reactive HLH/ MAS be considered sooner. However, this view emphasizes that any specific anti-inflammatory therapy, once initiated, needs to be tolerable, if not effective, for all the disorders under consideration. This report suggests, like others recently published, that initial therapy for secondary HLH may have shifted from primarily being myelosuppressive to being anti-inflammatory (2, 10, 11). Most experts would agree that critically ill patients who are unstable with secondary HLH might need more aggressive yet well-tolerated therapy than that offered by high-dose pulse steroids alone (8, 16). This study suggests that anakinra could be used to treat the inflammation associated with secondary HLH. In our study, anakinra was started only after patients were determined to have findings atypical for sepsis and after rheumatology and infectious disease consultation at median day 5 of the hospital course (range, 3–13) and administered for a median of 8.5 days. Anakinra seemed to be well tolerated and the one patient who developed pulmonary aspergillosis did so 214 days after anakinra therapy, while receiving reduced intensity chemotherapy in preparation for HSCT. Infections serve primarily as triggers and it is thought that functional deficiencies in NK cells and cytotoxic T cells that occur during the illness drive the response of the host to the infection (17). This cytokine surge, and not the original infection, characterize the systemic manifestation of secondary HLH and eventually lead to end-organ damage and mortality. More than any other cytokine family, the IL-1 family of ligands and receptors is associated with the inflammatory response (18). Studies have shown that the IL-1 receptor stimulation plays a key role in the evolution and persistence of end-organ injury and its down-regulation leads to improved cardiopulmonary function in animal models (19–22). Our study showed that patients had clinical improvement with less ventilator
Characteristics of Patients Admitted into the ICU and Treated With Anakinra
Patient
Infections
Diagnostic Source
Reason for ICU
1
Histoplasma capsulatum
Blood culture drawn at anakinra start Increased WOB Pulmonary/renal/hepatic/CVS
2
Mycobacterium avium complex BAL
Increased WOB Pulmonary/renal
3
None identified
Chest pain
CVS/hepatic
4
None identified
Hypotension
CVS/renal/hepatic
5
Varicella zoster
Skin lesion culture
Increased WOB Pulmonary/CVS/renal
6
Candida sphaerica
BAL
Hypotension
7
Epstein-Barr virus
Serum polymerase chain reaction
Increased WOB Pulmonary/hepatic
8
None identified
Hypotension
Systems Involved
CVS/gastrointestinal/pulmonary Cardiac/hepatic
WOB = work of breathing, CVS = cardiovascular system, BAL = bronchoalveolar lavage. a Included milrinone. b Patient received continuous renal replacement therapy that was transitioned to hemodialysis.
406
www.pccmjournal.org
June 2014 • Volume 15 • Number 5
Feature Articles
requirement and ultimately ICU discharge. However, this study is inadequately powered to establish the clinical benefits of IL-1 antagonism in patients with secondary HLH. In spite of this and other reports that tout the clinical benefits of IL-1 blockade, studies have not shown IL-1 levels to be consistently raised during HLH. Findings by Fujiwara et al (23) demonstrated elevated concentrations of IL-1 in only five of 27 patients with hemophagocytic syndromes and Ishii et al (9) observed such elevation in only five of 12 children with HLH. Henter et al (24) showed an increase in IL-1ra concentration, but not IL-1 levels. They hypothesized that it could be that IL-1 is secreted locally and hence not measured, but the resulting increase in antagonist IL-1ra levels may be a part of the systemic response. This may explain why IL-1 receptor blockade may offer a therapeutic role despite the inconsistency of measured levels of IL-1 in the above studies. In addition, there are data that favor a hypothesis suggesting some patients have a phenotype associated with a tendency to increased phagocyte activity. In these patients, the anti-inflammatory, negative feedback loop induced by innate immune responses to IL-1 and IL-6 may favor conditions leading to an alternative pathway of macrophage differentiation, and thus facilitate development of the so-called CD163 “scavenger” phenotype (25). This phenotype is associated with increased phagocytic activity and has been implicated in the development of hemophagocytic syndromes. In theory, blocking the activity of IL-1 should blunt this feedback loop. Infection remains a prominent cause of mortality and morbidity in patients with all forms of HLH. In a single-center retrospective study of 18 children identified with primary HLH, an infectious agent was documented at the initial presentation of HLH in five patients. Significant infections occurred during therapy in 10 of 18 patients (56%). Of the 12 fatal cases, invasive infection was the cause of death in eight children, and six of these deaths were attributable to invasive fungal infection (26). In our study, infections were present as triggers, but we were able to treat these infections while the patients continued to receive anakinra. In our report, mortality was lower, 1 of
Pediatric Risk of Mortality Score
8 (12.5%), when compared to 25–50% reported in previously published articles (27, 28). Studies have shown solid organ transplant patients who develop HLH to have high mortality (28, 29). It is notable that two patients in our cohort who received solid organ transplant survived. High-dose corticosteroid therapy is an important confounding factor to our study as this is considered first-line therapy for secondary HLH. It remains possible that our patients may have responded to steroids alone without the addition of anakinra. Stéphan et al (30) in their article on reactive HLH showed that IV steroids at doses ranging from conventional to pulse methylprednisolone induced remission in 15 of 21 episodes (71%) when used alone as the first-line treatment for reactive HLH. In that study, 10 patients (47%) needed ICU admission and it is unclear whether any of these patients failed to respond to initial therapy. Limitations of Study There are obvious limitations in a case series of this type. As is common with other reports on HLH, we have undertaken a small retrospective study with eight patients who presented with a wide variety of clinical manifestations and management approaches including corticosteroids. We have considered that some of the patients included in this study ultimately may not have had secondary HLH as confirmatory tissue diagnosis was lacking; however, there was a strong suggestion and partial fulfillment of sufficient diagnostic criteria to support the diagnosis of secondary HLH and justified a trial of aggressive therapy. The lack of consistent BM findings is an additional potential limitation. A recent study by Demirkol et al (5) used the presence of BM hemophagocytosis as sine qua non for diagnosis. In our case series, BMA showed significant hemophagocytosis in only two (cases 3 and 6) over eight. However, in diagnosing secondary HLH, the presence and extent of BM hemophagocytosis by itself is not specific or sensitive. One study that used a range of 1–10 hemophagocytes/500 cells still had only 58% of HLH patients with bone marrow aspirate suggestive of HLH and suggested a negative initial bone marrow aspirate should not deter diagnosis (31). Yet another showed
Chest Radiograph
Mechanical Ventilation Days
Vasoactive Infusions Use
Renal Replacement Therapy
Survival
3
Bilateral infiltrates
None
Yesa
No
Yes
0
Bilateral infiltrates
5
No
No
Yes
13
Pleural effusion
None
No
No
Yes
12
Pleural effusion
None
Yes
No
Yes
18
Bilateral infiltrates
5
Yes
a
Yes
Yes
17
Bilateral infiltrates
10
Yes
a
No
No
8
Bilateral infiltrates
6
No
No
Yes
11
Bilateral infiltrates
5
Yes
No
Yes
Pediatric Critical Care Medicine
a
b
www.ccmjournal.org
407
Rajasekaran et al
patients who did not have HLH but died of other critical illness to have significant BM hemophagocytosis on necropsy (32). Therefore, BM hemophagocytosis remains only one of the criteria for making the diagnosis of secondary HLH. We did not have any patient with significant side effects that needed discontinuation of anakinra.
CONCLUSIONS The 1994 recommendations by the Histiocyte society continue to be the backbone of initial therapy for the unresponsive or unstable patient with confirmed or suspected HLH in many institutions (1). Our experience using anakinra in these patients with confirmed or strongly suspected secondary HLH provides a promising foundation for further investigation of this agent as part of the therapeutic regimen, with the potential to safely improve the clinical outcome in patients with acquired, or secondary, forms of HLH. As most cases of secondary HLH are diagnosed in the ICU, it is essential that pediatric intensivists have knowledge of this entity in order to make an early diagnosis and promptly initiate appropriate therapy.
ACKNOWLEDGMENT We thank Jackson R. Whitman for his excellent support in the preparation of this article.
REFERENCES
1. Henter JI, Aricò M, Egeler RM, et al: HLH-94: A treatment protocol for hemophagocytic lymphohistiocytosis. HLH study Group of the Histiocyte Society. Med Pediatr Oncol 1997; 28:342–347 2. Janka GE: Familial and acquired hemophagocytic lymphohistiocytosis. Eur J Pediatr 2007; 166:95–109 3. 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 4. 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 5. Demirkol D, Yildizdas D, Bayrakci B, et al: Hyperferritinemia in the critically ill child with secondary HLH/sepsis/MODS/MAS: What is the treatment? Crit Care 2012; 16:R52 6. Risma K, Jordan MB: Hemophagocytic lymphohistiocytosis: Updates and evolving concepts. Curr Opin Pediatr 2012; 24:9–15 7. Henter JI, Horne A, Arico M, et al: HLH-2004: Diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer 2007; 48:124–131 8. Jordan MB, Allen CE, Weitzman S, et al: How I treat hemophagocytic lymphohistiocytosis. Blood 2011; 118:4041–4052 9. Ishii E, Ohga S, Aoki T, et al: Prognosis of children with v irus-associated hemophagocytic syndrome and malignant histiocytosis: Correlation with levels of serum interleukin-1 and tumor necrosis factor. Acta Haematol 1991; 85:93–99 10. Bruck N, Suttorp M, Kabus M, et al: Rapid and sustained remission of systemic juvenile idiopathic arthritis-associated macrophage activation syndrome through treatment with anakinra and corticosteroids. J Clin Rheumatol 2011; 17:23–27 11. Miettunen PM, Narendran A, Jayanthan A, et al: Successful treatment of severe paediatric rheumatic disease-associated macrophage activation syndrome with interleukin-1 inhibition
408
www.pccmjournal.org
following conventional immunosuppressive therapy: Case series with 12 patients. Rheumatology (Oxford) 2011; 50:417–419 12. Wilkinson JD, Pollack MM, Ruttimann UE, et al: Outcome of pediatric patients with multiple organ system failure. Crit Care Med 1986; 14:271–274 13. Proulx F, Fayon M, Farrell CA, et al: Epidemiology of sepsis and multiple organ dysfunction syndrome in children. Chest 1996; 109:1033–1037 14. Nigrovic PA, Mannion M, Prince FH, et al: Anakinra as first-line disease-modifying therapy in systemic juvenile idiopathic arthritis: Report of forty-six patients from an international multicenter series. Arthritis Rheum 2011; 63:545–555 15. Opal SM, Fisher CJ Jr, Dhainaut JF, et al: Confirmatory interleukin-1 receptor antagonist trial in severe sepsis: A phase III, randomized, double-blind, placebo-controlled, multicenter trial. The Interleukin-1 Receptor Antagonist Sepsis Investigator Group. Crit Care Med 1997; 25:1115–1124 16. Henter JI, Samuelsson-Horne A, Aricò M, et al; Histocyte Society: Treatment of hemophagocytic lymphohistiocytosis with HLH-94 immunochemotherapy and bone marrow transplantation. Blood 2002; 100:2367–2373 17. Janka GE: Familial and acquired hemophagocytic lymphohistiocytosis. Annu Rev Med 2012; 63:233–246 18. Dinarello CA: Interleukin-1 in the pathogenesis and treatment of inflammatory diseases. Blood 2011; 117:3720–3732 19. Goodman RB, Strieter RM, Martin DP, et al: Inflammatory cytokines in patients with persistence of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1996; 154:602–611 20. Hybertson BM, Lee YM, Repine JE: Phagocytes and acute lung injury: Dual roles for interleukin-1. Ann N Y Acad Sci 1997; 832:266–273 21. Meduri GU, Headley S, Kohler G, et al: Persistent elevation of inflammatory cytokines predicts a poor outcome in ARDS. Plasma IL-1 beta and IL-6 levels are consistent and efficient predictors of outcome over time. Chest 1995; 107:1062–1073 22. Pino-Yanes M, Ma SF, Sun X, et al: Interleukin-1 receptor-associated kinase 3 gene associates with susceptibility to acute lung injury. Am J Respir Cell Mol Biol 2011; 45:740–745 23. Fujiwara F, Hibi S, Imashuku S: Hypercytokinemia in hemophagocytic syndrome. Am J Pediatr Hematol Oncol 1993; 15:92–98 24. Henter JI, Andersson B, Elinder G, et al: Elevated circulating levels of interleukin-1 receptor antagonist but not IL-1 agonists in hemophagocytic lymphohistiocytosis. Med Pediatr Oncol 1996; 27:21–25 25. Fall N, Barnes M, Thornton S, et al: Gene expression profiling of peripheral blood from patients with untreated new-onset systemic juvenile idiopathic arthritis reveals molecular heterogeneity that may predict macrophage activation syndrome. Arthritis Rheum 2007; 56:3793–3804 26. Sung L, Weitzman SS, Petric M, et al: The role of infections in primary hemophagocytic lymphohistiocytosis: A case series and review of the literature. Clin Infect Dis 2001; 33:1644–1648 27. Créput C, Galicier L, Buyse S, et al: Understanding organ dysfunction in hemophagocytic lymphohistiocytosis. Intensive Care Med 2008; 34:1177–1187 28. Diaz-Guzman E, Dong B, Hobbs SB, et al: Hemophagocytic lymphohistiocytosis after lung transplant: Report of 2 cases and a literature review. Exp Clin Transplant 2011; 9:217–222 29. Karras A, Thervet E, Legendre C; Groupe Coopératif de transplantation d’Ile de France: Hemophagocytic syndrome in renal transplant recipients: Report of 17 cases and review of literature. Transplantation 2004; 77:238–243 3 0. Stéphan JL, Koné-Paut I, Galambrun C, et al: Reactive haemophagocytic syndrome in children with inflammatory disorders. A retrospective study of 24 patients. Rheumatology (Oxford) 2001; 40:1285–1292 31. Gupta A, Tyrrell P, Valani R, et al: The role of the initial bone marrow aspirate in the diagnosis of hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer 2008; 51:402–404 32. Strauss R, Neureiter D, Westenburger B, et al: Multifactorial risk analysis of bone marrow histiocytic hyperplasia with hemophagocytosis in critically ill medical patients—A postmortem clinicopathologic analysis. Crit Care Med 2004; 32:1316–1321 June 2014 • Volume 15 • Number 5
Therapeutic Role of Anakinra, an Interleukin-1 Receptor Antagonist, in the Management of Secondary Hemophagocytic Lymphohistiocytosis/ Sepsis/Multiple Organ Dysfunction/Macrophage Activating Syndrome in Critically Ill Children* Surender Rajasekaran, MD, MPH1; Katherine Kruse, MD1,2; Karen Kovey, PharmD1; Alan T. Davis, PhD2; Nabil E. Hassan, MD1; Akunne N. Ndika, MBBS, MPH1; Sandra Zuiderveen, BSN, RN1; James Birmingham, MD3
Objectives: Secondary hemophagocytic lymphohistiocytosis, macrophage activating syndrome, and sepsis share the same inflammatory phenotype leading often to multiple organ dysfunction syndrome needing intensive care. The goal of this article is to describe our experience with anakinra (Kineret), a recombinant interleukin-1 receptor antagonist, in decreasing the systemic inflammation. Design: Retrospective case series. Setting: The PICU at the Helen DeVos Children’s Hospital (Grand Rapids, MI). Patients: The records of eight critically ill children presumed to have secondary hemophagocytic lymphohistiocytosis at our institution between January 1, 2011, and July 31, 2012, were reviewed. Interventions: All of the patients were treated with anakinra (Kineret) and in some cases systemic corticosteroids as first-line therapy for secondary hemophagocytic lymphohistiocytosis. Measurements and Main Results: Patients had a median age of 14 years and a median Pediatric Risk of Mortality score of 11.5. Four were previously healthy and four had underlying diseases that could have made them susceptible to secondary hemophagocytic lymphohistiocytosis. Indications for PICU transfer were respiratory distress 50% (4 *See also p. 486. 1 Pediatric Critical Care Medicine, Helen DeVos Children's Hospital. Grand Rapids, MI. 2 Grand Rapids Medical Education Partners, Grand Rapids, MI. 3 Department of Rheumatology, Helen DeVos Children's Hospital. Grand Rapids, MI. The authors have disclosed that they do not have any potential conflicts of interest. For information regarding this article, E-mail: surender.rajasekaran@ helendevoschildrens.org Copyright © 2014 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies DOI: 10.1097/PCC.0000000000000078
Pediatric Critical Care Medicine
of 8), cardiovascular instability 37.5% (3 of 8), and chest pain (1 of 8). Five of the patients (62.5%) were mechanically ventilated and 62.5% (5 of 8) received vasoactive infusions. Inflammatory markers were assessed linearly at the start of therapy and 7 days later. Baseline C-reactive protein was 206 ± 50 mg/L (mean ± sem) at the start of anakinra and decreased by 67.1% to 68 ± 36 mg/L (p = 0.03). Ferritin decreased by 63.8% to 3,210 ± 1,178 ng/mL (p = 0.30), and fibrinogen decreased by 42% to 158 ± 41 mg/dL (p = 0.03). Absolute neutrophil count (p = 0.38) and absolute lymphocyte count (p = 0.69) did not change significantly. No infections were attributed to anakinra therapy. One patient died long after treatment with anakinra while receiving pre-hematopoietic stem cell transplant chemotherapy. Conclusions: Anakinra could represent a promising therapeutic approach in these life-threatening disorders that are likely underdiagnosed and often difficult to treat. (Pediatr Crit Care Med 2014; 15:401–408) Key Words: cytokines; hemophagocytic lymphohistiocytosis; inflammation; macrophage activating syndrome; multiple organ dysfunction syndrome; pharmacotherapy
F
or the last 2 decades, the unstable patient with hemophagocytic lymphohistiocytosis (HLH) has been treated using systemic corticosteroids and etoposide in most centers. This regimen, developed in 1994 by the histiocyte society through expert consensus, continues to be used despite its many known side effects (1). The society has broadened its diagnostic criteria in order to make an earlier diagnosis and it is now recognized that significant clinical and laboratory overlap occurs between secondary HLH/macrophage activating syndrome (MAS) and sepsis/systemic inflammatory response syndrome (SIRS)/multiple organ dysfunction syndrome www.pccmjournal.org
401
Rajasekaran et al
(MODS) (2–5). The risk of applying any HLH/MAS-targeted therapies to children with sepsis/SIRS/MODS has to be considered whatever therapeutic modality is eventually chosen, as patients with sepsis/SIRS might suffer harm from a therapy that targets inflammation (3). Therefore, it is critical that the risks of treatment or nontreatment be weighed according to the clinical presentation of each patient. As such it would be helpful if the therapy is tolerated by all the grouped disorders. Patients with HLH spectrum of disease often present with nonspecific signs and symptoms that mimic viral or bacterial illness. For the intensivist, it is important to understand the diagnostic criteria and management strategies because the vast majority of these HLH diagnoses are made in the PICU (6). Early diagnosis remains challenging and is often subjective and should be made in consultation with other subspecialties. The primary diagnosis is based on a series of predetermined clinical and laboratory criteria derived from the HLH-2004 consensus (7). It is recommended that initial therapy be the same regardless of HLH type (8). Nevertheless, an understanding of the various types of HLH may assist in developing management strategies and understanding prognosis. The Subtypes of HLH Patients in the “primary” HLH category are those with familial inheritance or genetic causes, are usually younger, and are
thought to have fixed defects of cytotoxic cell function. They have a clear risk of HLH recurrence and are not likely to survive without hematopoietic stem cell transplantation (HSCT). Although HLH in these patients can also be associated with infections, the immunologic trigger may not be apparent. The term “secondary HLH” or “reactive HLH” generally refers to older children who present without a family history or known genetic cause for their HLH. These patients typically have concurrent infections or chronic illness that appears to trigger the HLH (8). MAS is a term typically used to describe HLH associated with rheumatologic disorders but is likely not pathophysiologically distinct from secondary or reactive forms of HLH. For the purposes of our discussion, we will use the term “secondary HLH” to encompass the spectrum disorders under our consideration for early identification and management. Therapy is directed at ameliorating inflammation by attenuating the hypercytokinemia (or what has been termed the “cytokine storm”) while simultaneously supporting failing organs. Steroids, chemotherapy, and stem cell transplantation are treatment options, but these often lead to opportunistic infections and further organ damage. The clinical and laboratory findings in HLH can be anticipated based on the known biological effects of the inflammatory cytokines tumor necrosis factor-α and interleukin (IL)-1. Some studies have shown elevated serum IL-1 in patients with HLH (9). As would be
Table 1. Patient Characteristics and Laboratory Values Performed Within 72 Hours of Anakinra Initiation Patient
1
Age (yr)/Sex
15/M
Preexisting Conditions
Hyper IgM CD40 ligand deficiency
Organ/Lymph Node
Bone Marrow/Tissue Phagocytes
Natural Killer Activity
Ferritin Prior (ng/mL)
Liver
None
Not done
19,790
Spleen LN 2
20/M
Healthy
None
+
Low
23,320
3
13/M
Healthy
Liver
++
Low
1,138
Lymph node
Normal
623
Spleen
No bone marrow done
Low
683
LN
Peripheral cytometry
LN 4
15/F
Family history of systemic lupus erythematosus
Liver Spleen LN
5
21/F
Renal transplant 2009 lupus variant
6
11/M
Healthy
None
++
Low
28,931
7
8/M
Liver transplant acetyl CoA deficiency
Liver
+
Not done
1,372
+
Not done
6,651
Spleen 8
9/F
Healthy
Liver Spleen LN
LN = lymphadenopathy, + = few hemophagocytic histiocytes, ++ = frequent hemophagocytic histiocytes. a Triglyceride the highest from the first 72 hr of therapy.
402
www.pccmjournal.org
June 2014 • Volume 15 • Number 5
Feature Articles
hypothesized, several case reports and small series have demonstrated benefits of IL-1 antagonist therapy in patients with secondary HLH. However, these have been either single patient reports or one small series involving patients with juvenile idiopathic arthritis (JIA) that developed MAS (10, 11). In this article, we examine the clinical course of eight critically ill pediatric patients with presumed or confirmed secondary HLH who were treated with anakinra.
METHODS Following approval by the institutional review board, we performed a retrospective review of eight patients with suspected secondary HLH who received care in our PICU at Helen DeVos Children’s Hospital between January 2011 and July 2012. These patients were ultimately treated with anakinra as part of their management. A retrospective review of the Cerner database was conducted to identify all patients who were worked up for or identified as having secondary HLH. Once identified, their clinical course was plotted prior to and after initiation of anakinra. Virtual PICU systems and Cerner were used to determine the clinical course and laboratory indices. Patients included in our analysis were PICU patients diagnosed with secondary HLH and received anakinra as part of the therapy. Patients with secondary HLH not requiring PICU admission were excluded from our review. Various clinical and
Ferritin Highest (ng/mL)
Cytopenia > 2 Cell Lines
Elevated Liver Enzymes
No
Yes
57
287
12,000
No
No
No
114
174
1,207
No
4,757 (day 4)
No
Yes
238
369
20,000
No
1,552 (day 3)
Yes
Yes
186
258
3,066
No
1,100 (day 3)
Yes
Yes
134
322a
11,128
No
No
No
44
343
2,213
No
Yes
Yes
76
168a
12,220
Yes
No
Yes
415
25,360 (day 3)
Pediatric Critical Care Medicine
Fibrinogen (mg/dL)
laboratory findings were extracted from the specified databases and presented in the Results section. Typically, the patients with suspected HLH are managed in consultation with infectious disease, rheumatology, and hematology providers. The diagnosis was ultimately decided in conjunction with evaluation by the pediatric rheumatologist who then determined when to initiate anakinra. Numerous factors were considered in the determination of diagnosis and the timing of starting therapy with anakinra; however, several factors were universal including fever, laboratory and clinical evidence of uncontrolled inflammation (hyperferritinemia was an absolute requirement), lack of a definitive infectious source, and evidence of clinical deterioration. The diagnosis of secondary HLH was made with consideration of the criteria established by the HLH-2004 consensus (7) that identified the following criteria: familial history or known genetic defect and clinical/laboratory criteria including fever, splenomegaly, cytopenia ≥ 2 cell lines, hemoglobin < 9 g/dL, platelets < 100,000/L, neutropenia < 1 × 109/L, triglyceridemia, ferritin ≥ 500 ng/mL, fibrinogen < 150 mg/ dL, soluble CD25, decreased natural killer (NK) cell activity, hemophagocytosis in bone marrow (BM), lymph node, and cerebrospinal fluid (CSF), and elevated transaminases, bilirubin, lactate dehydrogenase (LDH), CSF, pleocytosis, or protein (7). Organ system dysfunction was defined according to the criteria established by Wilkinson et al (12) and subsequently
Triglyceride (mg/dL)
246
Lactate Dehydrogenase (IU/L)
Cerebrospinal Fluid Pleocytosis
No
www.ccmjournal.org
403
Rajasekaran et al
modified by Proulx et al (13). BM or other tissue confirmation was attempted in all patients, but incomplete or partial fulfillment of the criteria was deemed sufficient for therapeutic trial of anakinra when an alternate diagnostic explanation for the individual’s clinical status was not identified. Dosing of Anakinra Anakinra (Kineret) is available in single-use 1-mL prefilled glass syringes for daily subcutaneous (SC) administration. Each 1-mL prefilled glass syringe contained 0.67 mL (100 mg) of anakinra in a solution (pH 6.5) containing sodium citrate (1.29 mg), sodium chloride (5.48 mg), disodium EDTA (0.12 mg), and polysorbate 80 (0.70 mg) in water for injection, USP. Dosing in our patients was determined by the prescribing rheumatologist. The dose was initially based on patient weight but was rapidly increased based on evaluation of the clinical response. It should be noted that the usual recommended dose of anakinra (usually between 1 and 2 mg/kg/d) is for patients with JIA and is not necessarily applicable in this setting. We were comfortable using a higher than commonly recommended doses of anakinra for several reasons. First, the half-life of anakinra SC is very short (4–6 hr), justifying more frequent than daily dosing in the setting of acute systemic inflammation, in order to rapidly achieve therapeutic IL-1 inhibition. Additionally, prior experience with systemic juvenile arthritis patients had shown that children often require higher doses, as much as 11 mg/kg/d (14). Finally, there is published experience using high doses of continuous IV IL-1 receptor antagonist (IL-1ra) in sepsis patients, providing support for the safety of higher doses in an SIRS setting, even without evidence of
Figure 1. Clinical course of the patients.
404
www.pccmjournal.org
any compelling therapeutic effect in that analysis (15). Dosing was maintained until serological markers of inflammation were noted to be declining and this correlated with clinical improvement, further adjustments in the dosing of anakinra were determined by the rheumatologist. In addition to above therapy, five patients received IV immune globulin (IVIG) therapy and six patients received high-dose corticosteroid therapy with doses ranging from 10 to 20 mg/kg/d. Kineret prescribing information is available at http:// www.accessdata.fda.gov/drugsatfda_docs/label/2003/anakamg062703LB.pdf. Statistical Methods Summary statistics were calculated. Quantitative data are expressed as the mean ± sem, whereas nominal data are expressed as a percentage. For various laboratory indices, comparisons between the time period just prior to administration of anakinra (anakinra start) and 1 week postadministration were performed using the Wilcoxon signed rank test. Significance was assessed at p less than 0.05. Analyses were performed using SPSS v.21 (IBM, Armonk, NY).
RESULTS Eight patients admitted to the PICU for HLH from January 2011 to July 2012 were included for analysis. Five of the eight patients were mechanically ventilated for respiratory failure and five patients needed vasoactive infusions to maintain cardiovascular function. One patient needed renal replacement therapy (RRT) and another with the most prolonged course died 217 days post PICU admission. Patient Characteristics Patient characteristics, including most of the clinical and laboratory features used to make the initial diagnosis of secondary HLH, are described in Table 1. Eight patients (five males and three females) with an average age of 13.8 ± 1.6 years were diagnosed with secondary HLH. Four patients had preexisting diseases and four were previously healthy. Six patients had bone marrow aspiration (BMA) done with only two demonstrating significant hemophagocytosis and three showing some phagocytosis. Patient 4 did have a lymph node examination during the admission which was June 2014 • Volume 15 • Number 5
Feature Articles
Diagnostic Findings of Patients With Hyperferritinemia and Secondary Hemophagocytic Lymphohistiocytosis/ Sepsis/Multiple Organ Dysfunction Syndrome/Macrophage Activating Syndrome
making the initial assessment of secondary HLH (Table 1). Later, ferritin and CRP levels were used to evaluate therapy. Six of the patients received steroids with anakinra and five received IVIG. The dosing and duration of anakinra were determined as noted earlier. Anakinra was started at median day 5 of hospitalization (range, 3–13) and median duration of anakinra administration was 8.5 days (range, 5–31 d; Fig. 1). The single mortality was ultimately suspected of having primary HLH (no perforin 1 mutations detected) or at least a more aggressive form of HLH. Therapy was changed to the typical HLH protocol including etoposide within 2 weeks of discontinuing anakinra as his symptoms quickly recurred. By the time of death, he had been off anakinra for months (Fig. 1).
Table 2.
Secondary Hemophagocytic Lymphohistio cytosis/Macrophage Activating Syn drome (%) (n = 8)
Variable
Fever
100 (8)
Lactate dehydrogenase > 1,000 IU/L
100 (8)
Hyperferritinemia > 500 ng/mL
100 (8)
Bicytopenia
87 (7)
Anemia < 9 g/dL
87 (7)
Survival
87 (7)
Thrombocytopenia platelet < 100/μL
62 (5)
Hypertriglyceridemia > 265 mg/dL
62 (5)
Hemophagocytosis in bone marrow
62 (5)
Splenomegaly
50 (4)
Hypofibrinogenemia < 150 mg/dL
50 (4)
Neutropenia absolute neutrophil count < 1,000/μL
25 (2)
Markers of Inflammation and Other Laboratory Indices The clinical markers suggestive of secondary HLH were found in varying degrees with fever, elevated ferritin (> 500 ng/mL) and elevated LDH (> 1,000 IU/L) being found in 100% of patients (Table 2). The concentrations of a variety of markers of inflammation at various time points before and after anakinra administration are described in Table 3. All seven of the markers listed showed declines in their values from just prior to administration of anakinra (anakinra start) to 1 week after administration, although this change was only statistically significant for CRP and fibrinogen (p = 0.03). Mean serum ferritin decreased by 63.8% during this time period (p = 0.30), absolute neutrophil count (ANC) decreased by 30.2% (p = 0.38), and platelets by 29.3% (p > 0.99).
All of the patients met five or more of the criteria required for a diagnosis of hemophagocytic lymphohistiocytosis/macrophage activating syndrome.
not typical for HLH. Patient 5 did not receive a BMA as the attending on service felt the procedure could compromise the vascular access that was necessary for RRT in this fluid overloaded patient. Five patients had peripheral NK cells examined for perforin and granzyme B done for clinical reasons. The levels of serum triglyceride, ferritin, and C-reactive protein (CRP) in the first 72 hours of therapy were noted, as they assisted in Table 3.
Patients’ ICU Characteristics The patients had a median Pediatric Risk of Mortality score of 11.5 and spent an average of 9.3 ± 1.7 days in the PICU (Table 4). Patients 3, 4, and 8 were directly admitted into the ICU from out-of-hospital referring centers. Five patients had multiple organ involvement 48 hours into ICU admission necessitating vasoactive infusions, five needed ventilatory support, and one required RRT.
Laboratory Markers in Patients Treated With Anakinra
Variables
Admission
C-reactive protein (mg/L)
150 ± 28
72 Hr Post
1 Wk
2 Wk
206 ± 50a
98 ± 37
68 ± 36a
30 ± 9
8,098 ± 4,696
9,761 ± 4,212
7,300 ± 3,134
3,210 ± 1,178
1,464 ± 425
Fibrinogen (mg/dL)
508 ± 105
271 ± 108b
119 ± 32
158 ± 41b
267 ± 86
Creatinine (mg/dL)
0.7 ± 0.2
0.9 ± 0.3
1.0 ± 0.3
0.7 ± 0.1
0.6 ± 0.2
Lactate dehydrogenase (IU/L)
17,747 ± 17,045
2,593 ± 1,552
1,458 ± 690
994 ± 463
271 ± 59
Absolute neutrophil count (103/μL)
8,843 ± 5,550
6,854 ± 2,384
7,407 ± 2,356
4,786 ± 1,724
1,023 ± 494
229 ± 111
289 ± 103
327 ± 116
204 ± 60
235 ± 58
Ferritin (ng/mL)
Platelets (103/μL)
Anakinra Start
C-reactive protein showed statistically significant declines at 1 wk, in comparison to the values at anakinra start (p = 0.03). b Fibrinogen showed statistically significant declines at 1 wk, in comparison to the values at anakinra start (p = 0.03). None of the other variables showed significant changes for the same time interval (p > 0.05). Only three patients had complete blood counts at 2 wk. Values reported as the mean ± sem. a
Pediatric Critical Care Medicine
www.ccmjournal.org
405
Rajasekaran et al
The vasoactive infusions used to maintain perfusion included three patients receiving norepinephrine and milrinone, one receiving only milrinone, and one receiving dopamine. Mechanical ventilation was for a median of 6 days (range, 2–11 d; Fig. 1), and one patient (patient 3) received 2 L/min nasal cannula oxygen. All eight patients had chest radiograph changes (Table 4). The starting mean oxygenation index 12 hours after intubation and initiation of ventilation was 15.2. Initiation of anakinra was varied but the median start time was day 1 of mechanical ventilation (Fig. 1). One patient with a history of previous renal transplant needed to be on continuous RRT during her ICU course. Recovery in relation to anakinra administration is illustrated in Figure 1. Anakinra and Infection Patients 1, 2, 5, and 7 had infections around their anakinra course (Table 4). These infections were also possibly the triggers for the secondary HLH. Patient 6 had two infections: the first was 9 days after anakinra was stopped and the next one was prior to mortality. Patients were administered anakinra along with specific antimicrobials wherever indicated. Patient 1 had a histoplasma infection that was thought to be the trigger for secondary HLH, even though Figure 1 shows that the diagnosis occurred days into the anakinra course. At diagnosis, a single patient was neutropenic with 450 cells/μL. Later at 1 week, three patients became neutropenic (WBC < 1,500 cells/μL) and one patient became lymphopenic (absolute lymphocyte count [ALC] < 1,000 cells/μL) at day 7 of anakinra. Two of the three neutropenic patients were also on systemic steroids. The one patient who died had normal ANC and ALC during and immediately after his course of anakinra.
DISCUSSION Patients with secondary HLH present with the same phenotypical features as sepsis. Distinguishing them is important because the management of sepsis is one of intensive support Table 4.
while providing the appropriate antibiotic coverage whereas that for secondary HLH is to ameliorate inflammation. Considering these pathophysiologically related disorders together as sepsis/SIRS/MODS/MAS until a more specific diagnosis is made would encourage that the diagnosis of reactive HLH/ MAS be considered sooner. However, this view emphasizes that any specific anti-inflammatory therapy, once initiated, needs to be tolerable, if not effective, for all the disorders under consideration. This report suggests, like others recently published, that initial therapy for secondary HLH may have shifted from primarily being myelosuppressive to being anti-inflammatory (2, 10, 11). Most experts would agree that critically ill patients who are unstable with secondary HLH might need more aggressive yet well-tolerated therapy than that offered by high-dose pulse steroids alone (8, 16). This study suggests that anakinra could be used to treat the inflammation associated with secondary HLH. In our study, anakinra was started only after patients were determined to have findings atypical for sepsis and after rheumatology and infectious disease consultation at median day 5 of the hospital course (range, 3–13) and administered for a median of 8.5 days. Anakinra seemed to be well tolerated and the one patient who developed pulmonary aspergillosis did so 214 days after anakinra therapy, while receiving reduced intensity chemotherapy in preparation for HSCT. Infections serve primarily as triggers and it is thought that functional deficiencies in NK cells and cytotoxic T cells that occur during the illness drive the response of the host to the infection (17). This cytokine surge, and not the original infection, characterize the systemic manifestation of secondary HLH and eventually lead to end-organ damage and mortality. More than any other cytokine family, the IL-1 family of ligands and receptors is associated with the inflammatory response (18). Studies have shown that the IL-1 receptor stimulation plays a key role in the evolution and persistence of end-organ injury and its down-regulation leads to improved cardiopulmonary function in animal models (19–22). Our study showed that patients had clinical improvement with less ventilator
Characteristics of Patients Admitted into the ICU and Treated With Anakinra
Patient
Infections
Diagnostic Source
Reason for ICU
1
Histoplasma capsulatum
Blood culture drawn at anakinra start Increased WOB Pulmonary/renal/hepatic/CVS
2
Mycobacterium avium complex BAL
Increased WOB Pulmonary/renal
3
None identified
Chest pain
CVS/hepatic
4
None identified
Hypotension
CVS/renal/hepatic
5
Varicella zoster
Skin lesion culture
Increased WOB Pulmonary/CVS/renal
6
Candida sphaerica
BAL
Hypotension
7
Epstein-Barr virus
Serum polymerase chain reaction
Increased WOB Pulmonary/hepatic
8
None identified
Hypotension
Systems Involved
CVS/gastrointestinal/pulmonary Cardiac/hepatic
WOB = work of breathing, CVS = cardiovascular system, BAL = bronchoalveolar lavage. a Included milrinone. b Patient received continuous renal replacement therapy that was transitioned to hemodialysis.
406
www.pccmjournal.org
June 2014 • Volume 15 • Number 5
Feature Articles
requirement and ultimately ICU discharge. However, this study is inadequately powered to establish the clinical benefits of IL-1 antagonism in patients with secondary HLH. In spite of this and other reports that tout the clinical benefits of IL-1 blockade, studies have not shown IL-1 levels to be consistently raised during HLH. Findings by Fujiwara et al (23) demonstrated elevated concentrations of IL-1 in only five of 27 patients with hemophagocytic syndromes and Ishii et al (9) observed such elevation in only five of 12 children with HLH. Henter et al (24) showed an increase in IL-1ra concentration, but not IL-1 levels. They hypothesized that it could be that IL-1 is secreted locally and hence not measured, but the resulting increase in antagonist IL-1ra levels may be a part of the systemic response. This may explain why IL-1 receptor blockade may offer a therapeutic role despite the inconsistency of measured levels of IL-1 in the above studies. In addition, there are data that favor a hypothesis suggesting some patients have a phenotype associated with a tendency to increased phagocyte activity. In these patients, the anti-inflammatory, negative feedback loop induced by innate immune responses to IL-1 and IL-6 may favor conditions leading to an alternative pathway of macrophage differentiation, and thus facilitate development of the so-called CD163 “scavenger” phenotype (25). This phenotype is associated with increased phagocytic activity and has been implicated in the development of hemophagocytic syndromes. In theory, blocking the activity of IL-1 should blunt this feedback loop. Infection remains a prominent cause of mortality and morbidity in patients with all forms of HLH. In a single-center retrospective study of 18 children identified with primary HLH, an infectious agent was documented at the initial presentation of HLH in five patients. Significant infections occurred during therapy in 10 of 18 patients (56%). Of the 12 fatal cases, invasive infection was the cause of death in eight children, and six of these deaths were attributable to invasive fungal infection (26). In our study, infections were present as triggers, but we were able to treat these infections while the patients continued to receive anakinra. In our report, mortality was lower, 1 of
Pediatric Risk of Mortality Score
8 (12.5%), when compared to 25–50% reported in previously published articles (27, 28). Studies have shown solid organ transplant patients who develop HLH to have high mortality (28, 29). It is notable that two patients in our cohort who received solid organ transplant survived. High-dose corticosteroid therapy is an important confounding factor to our study as this is considered first-line therapy for secondary HLH. It remains possible that our patients may have responded to steroids alone without the addition of anakinra. Stéphan et al (30) in their article on reactive HLH showed that IV steroids at doses ranging from conventional to pulse methylprednisolone induced remission in 15 of 21 episodes (71%) when used alone as the first-line treatment for reactive HLH. In that study, 10 patients (47%) needed ICU admission and it is unclear whether any of these patients failed to respond to initial therapy. Limitations of Study There are obvious limitations in a case series of this type. As is common with other reports on HLH, we have undertaken a small retrospective study with eight patients who presented with a wide variety of clinical manifestations and management approaches including corticosteroids. We have considered that some of the patients included in this study ultimately may not have had secondary HLH as confirmatory tissue diagnosis was lacking; however, there was a strong suggestion and partial fulfillment of sufficient diagnostic criteria to support the diagnosis of secondary HLH and justified a trial of aggressive therapy. The lack of consistent BM findings is an additional potential limitation. A recent study by Demirkol et al (5) used the presence of BM hemophagocytosis as sine qua non for diagnosis. In our case series, BMA showed significant hemophagocytosis in only two (cases 3 and 6) over eight. However, in diagnosing secondary HLH, the presence and extent of BM hemophagocytosis by itself is not specific or sensitive. One study that used a range of 1–10 hemophagocytes/500 cells still had only 58% of HLH patients with bone marrow aspirate suggestive of HLH and suggested a negative initial bone marrow aspirate should not deter diagnosis (31). Yet another showed
Chest Radiograph
Mechanical Ventilation Days
Vasoactive Infusions Use
Renal Replacement Therapy
Survival
3
Bilateral infiltrates
None
Yesa
No
Yes
0
Bilateral infiltrates
5
No
No
Yes
13
Pleural effusion
None
No
No
Yes
12
Pleural effusion
None
Yes
No
Yes
18
Bilateral infiltrates
5
Yes
a
Yes
Yes
17
Bilateral infiltrates
10
Yes
a
No
No
8
Bilateral infiltrates
6
No
No
Yes
11
Bilateral infiltrates
5
Yes
No
Yes
Pediatric Critical Care Medicine
a
b
www.ccmjournal.org
407
Rajasekaran et al
patients who did not have HLH but died of other critical illness to have significant BM hemophagocytosis on necropsy (32). Therefore, BM hemophagocytosis remains only one of the criteria for making the diagnosis of secondary HLH. We did not have any patient with significant side effects that needed discontinuation of anakinra.
CONCLUSIONS The 1994 recommendations by the Histiocyte society continue to be the backbone of initial therapy for the unresponsive or unstable patient with confirmed or suspected HLH in many institutions (1). Our experience using anakinra in these patients with confirmed or strongly suspected secondary HLH provides a promising foundation for further investigation of this agent as part of the therapeutic regimen, with the potential to safely improve the clinical outcome in patients with acquired, or secondary, forms of HLH. As most cases of secondary HLH are diagnosed in the ICU, it is essential that pediatric intensivists have knowledge of this entity in order to make an early diagnosis and promptly initiate appropriate therapy.
ACKNOWLEDGMENT We thank Jackson R. Whitman for his excellent support in the preparation of this article.
REFERENCES
1. Henter JI, Aricò M, Egeler RM, et al: HLH-94: A treatment protocol for hemophagocytic lymphohistiocytosis. HLH study Group of the Histiocyte Society. Med Pediatr Oncol 1997; 28:342–347 2. Janka GE: Familial and acquired hemophagocytic lymphohistiocytosis. Eur J Pediatr 2007; 166:95–109 3. 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 4. 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 5. Demirkol D, Yildizdas D, Bayrakci B, et al: Hyperferritinemia in the critically ill child with secondary HLH/sepsis/MODS/MAS: What is the treatment? Crit Care 2012; 16:R52 6. Risma K, Jordan MB: Hemophagocytic lymphohistiocytosis: Updates and evolving concepts. Curr Opin Pediatr 2012; 24:9–15 7. Henter JI, Horne A, Arico M, et al: HLH-2004: Diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer 2007; 48:124–131 8. Jordan MB, Allen CE, Weitzman S, et al: How I treat hemophagocytic lymphohistiocytosis. Blood 2011; 118:4041–4052 9. Ishii E, Ohga S, Aoki T, et al: Prognosis of children with v irus-associated hemophagocytic syndrome and malignant histiocytosis: Correlation with levels of serum interleukin-1 and tumor necrosis factor. Acta Haematol 1991; 85:93–99 10. Bruck N, Suttorp M, Kabus M, et al: Rapid and sustained remission of systemic juvenile idiopathic arthritis-associated macrophage activation syndrome through treatment with anakinra and corticosteroids. J Clin Rheumatol 2011; 17:23–27 11. Miettunen PM, Narendran A, Jayanthan A, et al: Successful treatment of severe paediatric rheumatic disease-associated macrophage activation syndrome with interleukin-1 inhibition
408
www.pccmjournal.org
following conventional immunosuppressive therapy: Case series with 12 patients. Rheumatology (Oxford) 2011; 50:417–419 12. Wilkinson JD, Pollack MM, Ruttimann UE, et al: Outcome of pediatric patients with multiple organ system failure. Crit Care Med 1986; 14:271–274 13. Proulx F, Fayon M, Farrell CA, et al: Epidemiology of sepsis and multiple organ dysfunction syndrome in children. Chest 1996; 109:1033–1037 14. Nigrovic PA, Mannion M, Prince FH, et al: Anakinra as first-line disease-modifying therapy in systemic juvenile idiopathic arthritis: Report of forty-six patients from an international multicenter series. Arthritis Rheum 2011; 63:545–555 15. Opal SM, Fisher CJ Jr, Dhainaut JF, et al: Confirmatory interleukin-1 receptor antagonist trial in severe sepsis: A phase III, randomized, double-blind, placebo-controlled, multicenter trial. The Interleukin-1 Receptor Antagonist Sepsis Investigator Group. Crit Care Med 1997; 25:1115–1124 16. Henter JI, Samuelsson-Horne A, Aricò M, et al; Histocyte Society: Treatment of hemophagocytic lymphohistiocytosis with HLH-94 immunochemotherapy and bone marrow transplantation. Blood 2002; 100:2367–2373 17. Janka GE: Familial and acquired hemophagocytic lymphohistiocytosis. Annu Rev Med 2012; 63:233–246 18. Dinarello CA: Interleukin-1 in the pathogenesis and treatment of inflammatory diseases. Blood 2011; 117:3720–3732 19. Goodman RB, Strieter RM, Martin DP, et al: Inflammatory cytokines in patients with persistence of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1996; 154:602–611 20. Hybertson BM, Lee YM, Repine JE: Phagocytes and acute lung injury: Dual roles for interleukin-1. Ann N Y Acad Sci 1997; 832:266–273 21. Meduri GU, Headley S, Kohler G, et al: Persistent elevation of inflammatory cytokines predicts a poor outcome in ARDS. Plasma IL-1 beta and IL-6 levels are consistent and efficient predictors of outcome over time. Chest 1995; 107:1062–1073 22. Pino-Yanes M, Ma SF, Sun X, et al: Interleukin-1 receptor-associated kinase 3 gene associates with susceptibility to acute lung injury. Am J Respir Cell Mol Biol 2011; 45:740–745 23. Fujiwara F, Hibi S, Imashuku S: Hypercytokinemia in hemophagocytic syndrome. Am J Pediatr Hematol Oncol 1993; 15:92–98 24. Henter JI, Andersson B, Elinder G, et al: Elevated circulating levels of interleukin-1 receptor antagonist but not IL-1 agonists in hemophagocytic lymphohistiocytosis. Med Pediatr Oncol 1996; 27:21–25 25. Fall N, Barnes M, Thornton S, et al: Gene expression profiling of peripheral blood from patients with untreated new-onset systemic juvenile idiopathic arthritis reveals molecular heterogeneity that may predict macrophage activation syndrome. Arthritis Rheum 2007; 56:3793–3804 26. Sung L, Weitzman SS, Petric M, et al: The role of infections in primary hemophagocytic lymphohistiocytosis: A case series and review of the literature. Clin Infect Dis 2001; 33:1644–1648 27. Créput C, Galicier L, Buyse S, et al: Understanding organ dysfunction in hemophagocytic lymphohistiocytosis. Intensive Care Med 2008; 34:1177–1187 28. Diaz-Guzman E, Dong B, Hobbs SB, et al: Hemophagocytic lymphohistiocytosis after lung transplant: Report of 2 cases and a literature review. Exp Clin Transplant 2011; 9:217–222 29. Karras A, Thervet E, Legendre C; Groupe Coopératif de transplantation d’Ile de France: Hemophagocytic syndrome in renal transplant recipients: Report of 17 cases and review of literature. Transplantation 2004; 77:238–243 3 0. Stéphan JL, Koné-Paut I, Galambrun C, et al: Reactive haemophagocytic syndrome in children with inflammatory disorders. A retrospective study of 24 patients. Rheumatology (Oxford) 2001; 40:1285–1292 31. Gupta A, Tyrrell P, Valani R, et al: The role of the initial bone marrow aspirate in the diagnosis of hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer 2008; 51:402–404 32. Strauss R, Neureiter D, Westenburger B, et al: Multifactorial risk analysis of bone marrow histiocytic hyperplasia with hemophagocytosis in critically ill medical patients—A postmortem clinicopathologic analysis. Crit Care Med 2004; 32:1316–1321 June 2014 • Volume 15 • Number 5
