- Top of page
- PATIENTS AND METHODS
Reactive hemophagocytic syndrome (RHS) is characterized by the proliferation of benign hemophagocytic histiocytes in the bone marrow and the mononuclear phagocytic system. Risdall et al first described a group of patients with a distinct clinical syndrome including histiocytic hyperplasia, benign hemophagocytosis, and low blood cell counts (1). Subsequent to Risdall's original description, the full spectrum of RHS and the underlying diseases responsible for it have been recognized. RHS is often described as a complication of hematologic diseases, including acute leukemia, non-Hodgkin's lymphoma, and other illnesses involving immune deficiency [1–3]. Viral, bacterial, fungal, or parasitic infections may also be complicated by RHS (4). In acquired immunodeficiency syndrome, RHS may be associated with the infections that commonly affect patients (e.g., pneumococcal disease, pneumocystosis), Epstein-Barr virus infection, or malignancy (5–8).
RHS occurring in the course of adult systemic disease has rarely been reported. Immunosuppression induced by the systemic disease itself and exacerbated by immunosuppressive therapies may result in RHS, with or without an associated infection. The most frequently reported systemic diseases associated with RHS are systemic lupus erythematosus (SLE) and adult Still's disease (ASD) (9–15). Most previous articles have reported on a single case. Furthermore, the course and treatment of RHS occurring as a complication of systemic diseases has not so far been established clearly in the literature.
The aims of this study were 1) to describe RHS in adult systemic diseases in terms of its clinical manifestations, laboratory features, treatments, and outcomes; 2) to distinguish RHS induced by infection from that complicating the onset of systemic disease; and 3) to analyze the prognostic factors for RHS based on different clinical features and laboratory findings. We collected data on all French RHS cases occurring in the context of adult systemic diseases and put them in perspective with the results of a systematic review of RHS cases reported in the literature.
- Top of page
- PATIENTS AND METHODS
RHS is a frequently fatal condition that is generally associated with lymphoma or an infectious disease. It has rarely been described in the course of systemic diseases. In the largest published studies of RHS, systemic disease was found to be the underlying disorder in 2–5% of RHS cases (Table 4). RHS may occur as a rare but serious complication of systemic diseases; it has been reported in 6 of 250 (2.4%) cases of SLE (14) and in 1 of 23 (4%) cases of ASD (23).
Table 4. Cases of reactive hemophagocytic syndrome complicating adult systemic diseases in largest series of the literature
|Authors (ref)||Systemic disease (no.)||Cases of reactive hemophagocytic syndrome, no.||Mortality, no. (%)|
|Kaito et al (25)||Systemic lupus erythematosus (1), adult Still's disease (1)||34||20 (59)|
|Sailler et al (27)||Systemic lupus erythematosus (3)||99||49 (49)|
|Tsuda (2)||Systemic lupus erythematosus (1), dermatomyositis (1)||23||5 (22)|
|Wong and Chan (3)||Systemic lupus erythematosus (2)||40||18 (45)|
|Albert et al (24)||Systemic lupus erythematosus (2)||47||28 (62)|
|Risdall et al (1)||Systemic lupus erythematosus (1)||19||5 (26)|
|Reiner et al (26)||Systemic lupus erythematosus (2), rheumatoid arthritis (1), polymyalgia rheumatica (1)||23||7 (30)|
We report on 26 cases of RHS occurring in the course of systemic disease. The clinical and biologic features did not differ greatly from the classic description of RHS reported in previous studies (1, 2, 24–27). We show that the respective incidences of fever, hepatomegaly, splenomegaly, and lymphadenopathy were lower in our patients than in previous reports of RHS (24). One explanation for these findings is that the retrospective inclusion criteria used in our study may have selected clinically attenuated cases of RHS.
Our Medline search identified 40 well-documented cases of RHS occurring in the course of systemic disease. We found 19 cases of SLE (2, 12, 14, 26, 28–30), 9 cases of ASD (9–11, 13, 15, 31, 32), 5 cases of RA (26, 33–36), 1 case of Crohn's disease (26), 1 case of polymyalgia rheumatica (26), 2 cases of sarcoidosis (26, 37), 1 case of dermatomyositis (38), 1 case of systemic sclerosis (39), and 1 case of mixed connective tissue disease (39). The average age of the patients was 49.9 ± 15.0 years and the sex ratio was 30 women to 10 men. In the 27 cases where detailed data were available, RHS and the systemic disease appeared simultaneously on 11 occasions (the time elapsing being less than 1 month). In the other cases, systemic disease had been present for an average of 37 months before RHS was diagnosed. Ferritin concentrations were higher than 1,000 μg/liter in 13 cases and normal in 2 other cases (results only available for 15 cases). Higher ferritin levels were observed in cases of ASD (the mean ferritin concentration in ASD being 22,500 ± 2,790 μg/liter). Coagulopathy was observed in only 2 cases (results available for 21 cases). The clinical and biologic features of our patients did not differ from the RHS cases associated with systemic disease that had previously been reported in the literature (Table 5).
Table 5. Comparison between 26 cases of reactive hemophagocytic syndrome in systemic disease and 40 isolated, well-documented cases of the literature
| ||French cases (n = 26)||Literature cases (n = 40)|
|Age, mean ± SD, years||47.4 ± 17.7||49.9 ± 15.0|
|Sex ratio, F:M||19:7||30:10|
|Systemic lupus erythematosus, %||54||47|
|Onset of systemic disease, %||58||70|
This is the largest study to date on RHS occurring in the course of systemic disease, and it suggests that it is more common than previously recognized and represented in the literature. SLE and ASD predominated among the systemic diseases complicated by RHS.
RHS involves the proliferation of benign hemophagocytic histiocytes in the bone marrow and reticuloendothelial system. We showed that bone marrow aspirate was insufficient to obtain a diagnosis of RHS. Both bone marrow aspirate and biopsy are required, since some patients had a normal aspirate and a pathologic biopsy, whereas others had a normal biopsy and pathologic aspirate.
It appears that RHS may be associated with several different clinical settings, including infection, the onset of a systemic disease, or both.
In published cases of RHS occurring in patients with systemic disease, RHS was associated with the onset of disease and infection in 70% and 47% of cases, respectively (calculated from the data available on 34 cases).
An important finding in our study was that in SLE and ASD, the onset of disease alone seemed more likely to induce RHS. In cases of SLE and ASD, RHS was found to be associated with the onset of the disease in 72% of the cases. In contrast, with the other systemic diseases, infection was always present when RHS appeared. In the case of patients with RA, polyarteritis nodosa, systemic sclerosis, mixed tissue connective disease, or sarcoidosis, onset of the systemic disease alone was never found to be responsible for RHS.
This difference in the trigger disorders associated with RHS suggests that in SLE or ASD, infection and the disease should be treated together, whereas in other systemic diseases, infection should be treated and the immunosuppressive treatment of the underlying disease decreased.
In previous studies, the overall mortality associated with RHS was reported to be 46% (range 22–60%); however, these studies did not report specific mortality associated with RHS occurring in course of systemic diseases (Table 4). The prognostic factors associated with mortality due to RHS were studied in 34 patients (25), in whom the case fatality rate was 59%. In this study, factors associated with a higher risk of mortality were low hemoglobin levels, a low platelet count, high alkaline phosphatase, high bilirubin, high ferritin concentrations, and older age.
Our data suggest that the prognostic factors for RHS in systemic disease were thrombocytopenia, absence of lymphadenopathy, and the presence of corticosteroid therapy. SLE was not associated with a more favorable prognosis. Other factors, including the ferritin and LDH concentrations, were not significantly correlated with prognosis. However, it should be noted that because of the number of patients available, we had limited power to detect any significant associations.
The treatment strategies for RHS are not well established. The clinical setting and a set of prognostic factors need to be considered when choosing the optimum treatment strategy. In particular, when RHS is associated with an active infection occurring in an immunocompromised patient, the infection should be treated promptly and the immunosuppressive therapies decreased as much as possible. The majority of our patients were receiving immunosuppressive therapy when RHS occurred. In systemic diseases other than SLE and ASD, infection was the main factor for RHS, and onset of the disease alone was never associated with RHS. Therefore, treatment with antiinfectious agents and a reduction in immunosuppressive therapies should systematically be ensured in cases of RHS associated with RA, polyarteritis nodosa, mixed connective tissue disease, systemic sclerosis, or sarcoidosis. On the other hand, when RHS is associated with a systemic disease alone (a situation observed only in SLE and ASD), immunosuppressive therapy should probably be increased. In cases where both infection and ongoing systemic disease are associated with RHS, intravenous immunoglobulins in addition to antiinfectious agents may be considered. Immunoglobulin therapy has been shown to improve the outcome of a few pediatric cases of RHS (37, 40–42). Most of these patients were immunocompromised. In these cases, a clinical improvement was observed within 24–72 hours (42). Immunoglobulins may control both the viral replication and lymphohistiocytic dysregulation induced by infection. Only a prospective study will be able to determine the efficacy of combined therapy using antiinfectious agents and immunoglobulins in RHS. More generally, further studies are required to confirm the value of the treatment strategies proposed herewith for RHS associated with systemic disease.
RHS occurring in the course of systemic disease is an infrequent but important clinical entity in terms of patient prognosis. Physicians must keep the symptoms of RHS in mind, particularly because it is difficult to distinguish RHS symptoms from those of several systemic diseases. The treatment of RHS is an urgent matter and the optimum treatment strategy will depend on the underlying disorder: infection or the onset of systemic disease. In SLE and ASD, onset of the disease may be the only factor triggering RHS, whereas in other systemic diseases, infection (secondary to immunosuppression) is always present.