Abstract

Objective. The hyperimmunoglobulinaemia D and periodic fever syndrome (HIDS) was originally defined by the presence of a high serum level of immunoglobulin D associated with recurrent fever. Since the discovery of the mevalonate kinase gene (MVK) gene encoding the mevalonate kinase enzyme, most patients with a clinical diagnostic of HIDS are now found to have a mevalonate kinase deficiency based on metabolic and genetic data. We aimed to asses the value of a high IgD serum level for the diagnosis of HIDS in a cohort of patients with a phenotype of recurrent fever, and to characterize patients with a high IgD serum level without mevalonate kinase mutation.

Methods. Main clinical and biological data of 50 patients who presented with clinical signs compatible with HIDS have been prospectively registered on a standard form. Clinical data have been analysed according the IgD serum level and the presence of MVK mutation.

Results. The metabolic and genetic data establishing the diagnosis of HIDS correlated in all cases. In this series of 50 patients, the sensitivity of a high IgD value for the diagnosis of HIDS is 0.79. In five patients with MVK mutation, IgD levels were found to be in the normal range. Likelihood ratios indicate that IgD measurement is not relevant for the diagnostic of HIDS. Most patients with a high serum IgD level and no MVK mutation have no definite diagnosis.

Conclusion. The clinical relevance of the IgD measurement for the diagnosis of MKD in our population appears as poor, as reflected by likelihood ratios which are both close to 1.

Introduction

The hyperimmunoglobulinaemia D and periodic fever syndrome (HIDS) is a rare autosomal recessive disease, which belongs to the group of auto-inflammatory syndromes, characterized by febrile attacks recurring at more and less regular intervals.

This disease was originally described in patients of Dutch ancestry by Van der Meer et al. [1] in 1984 and was distinguished from familial Mediterranean disease (FMF) by the presence of a high serum level of immunoglobulin D. The clinical features of febrile attacks include lymphadenopathy, abdominal pain, diarrhoea, splenomegaly, hepatomegaly, skin rash, arthralgia and arthritis [2, 3]. Amyloidosis has for a long time considered to be absent in HIDS, but has been recently found in some patients, emphasizing the need to an early and thorough diagnosis of the disease [4, 5].

In 1999, mutations in mevalonate kinase gene (MVK), the gene which was previously known to underline mevalonic aciduria, were identified in HIDS patients [6, 7]. These mutations induce a variable reduction in the activity of mevalonate kinase (MK), an intermediary enzyme in cholesterol metabolism and isoprenoid biosynthesis. A complete mevalonate kinase deficiency (MKD) thus results in the mevalonic aciduria phenotype, and a partial one in the HIDS phenotype [8]. The molecular mechanisms by which defective activity of MVK may lead to fever attacks are still incompletely understood.

Until the discovery of the MVK gene as underlying HIDS, the presence of a high level of serum IgD was necessary to establish the diagnosis of HIDS, but since, few cases of HIDS with MVK mutations (i.e. MKD) and normal IgD serum levels have been described in the literature [7, 9, 10]. All of them were diagnosed during childhood. In addition, an elevated serum IgD is not a specific marker for HIDS.

The aim of this study was to evaluate the value of a high IgD serum level for the diagnosis of MKD and to characterize patients with a phenotype of recurrent fever syndrome with a high IgD serum level without MKD.

Patients and methods

Patients

The present study included patients referred to the Cochin hospital biochemistry and genetic molecular laboratory at Paris, a referral centre for the molecular diagnosis of periodic-fever syndromes. In this laboratory, routine molecular diagnosis began in 1997 for familial Mediterranean fever, in 1999 for tumor necrosis factor receptor-associated periodic syndrome (TRAPS) and HIDS and in 2001 for CIAS1 associated disorders.

Main clinical and biological data (age, sex, origin, family history, age at onset of inflammatory attacks, duration and frequency of attacks, organ involvement, abdominal pain, thoracic pain, arthralgias/arthritis, myalgias, skin rash and its type, eye inflammation and its type, deafness, splenomegaly, lymphadenopathydysmorphic features, IgD and IgA levels, C-reactive protein, treatments and their efficacy) have been prospectively registered on a standard form for all patients with a suspicion of hereditary periodic fever syndrome. In this study, we included all patients who presented with clinical signs compatible with HIDS. Neither specific clinical criteria nor a high IgD level was required to perform enzymatic activity and mutation research in the genomic DNA. From 1999 to June 2006, 144 patient's samples were searched for mutations in the MVK. Patients without any known IgD serum level were excluded as well as patients with the mevalonic aciduria phenotype defined by a very early onset of inflammatory attacks, growth and mental retardation and dysmorphic features. Fifty patients were thus included in our study.

Hyper IgD syndrome (HIDS) was defined in our study by clinical manifestations compatible with the diagnosis, and the presence of relevant MVK mutations. In our laboratory, the routine strategy is to search for mutations according to the clinical presentation in a step by step manner. Thus, for patients with a clinical suspicion of HIDS in the absence of MVK mutations, MEFV, TNFRSF1A and CIAS1 genes were studied according to the clinical data. Also, we communicated with all the clinicians in charge of the patients without a definite molecular diagnosis, in order to determine if a diagnosis was established prospectively.

Informed consent was obtained from all participants.

Analysis of mevalonate kinase gene mutations and enzyme activity and urinary mevalonic acid

Genomic DNA was extracted from whole blood or Epstein-Barr-immortalized lymphoblasts from the patients, and all the exons and intron/exons junctions of the MVK gene were amplified and sequenced according to established protocols [6, 8]. Mevalonate kinase enzyme activity was determined in peripheral blood lymphocytes using a radiometric assay as described previously [11, 12]. Mevalonic acid concentrations were measured by isotope-dilution gas chromatography–mass spectrometry from urines exclusively collected during a febrile attack [13].

Analysis of IgD serum level

IgD serum levels were measured on two occasions in patients, and one or more during attacks. The measure of serum IgD was done in three laboratories using the same radial immunodiffusion assay (Dade Behring). A high IgD serum was defined by a level >200 mg/l.

Statistical analysis

We calculated the specificity, sensitivity, predictive values and likelihood ratios of a high IgD serum level for the diagnosis of HIDS.

Unvaried statistical analysis was performed using chi-squared or Fisher exact tests for nominal variable, and Mann–Whitney tests for quantitative variables. We compared respectively patients with and without high level of serum IgD, and patients with high value of IgD with and without MVK mutation, performed on Stat View. The level of significance was set at P < 0.05.

Results

Fifty patients were included in our series. Twenty-four patients presented a MVK mutation, 19 with a high IgD serum level (sensitivity = 0.79).

Among the 38 patients with a high serum IgD, 19 were homozygotes or composite heterozygotes for MVK mutations and 19 did not bear any mutation in the coding part or in the intron–exon boundaries of the gene (positive predictive value = 0.5).

Twenty-six patients did not bear any mutation in MVK gene, 19 presenting a high serum IgD level (specificity = 0.27). The values are summarized in Table 1.

Table 1.

Sensitivity, specificity, predictive values and likelihood ratios of a high IgD serum concentration

 MKD + MKD− Total 
High level serum IgD 19 19 38 
Normal level IgD 12 
Total 24 26 50 
 MKD + MKD− Total 
High level serum IgD 19 19 38 
Normal level IgD 12 
Total 24 26 50 

MKD +, Presence of 2 mutations in MVK; MKD−, Absence of 2 mutations in MVK; Sensitivity, 0.79 [0.63–0.95]; Specificity, 0.27 [0.1–0.44]; Positive predictive value (VPP), 0.50 [0.34–0.66]; Negative predictive value (VPN), 0.58 [0.3–0.86]; Positive likelihood ratio, 1.08; Negative likelihood ratio, 0.7.

For most patients, mevalonate kinase activity measurement and/or mevalonate excretion during crisis had been performed. Comparison of the data revealed no discrepancy between biochemical tests and genetic results.

Characteristics of patients with high level IgD

Patients with MVK mutation were younger at the first attack (P = 0.0006), had more cutaneous involvement, lymphadenopathy (P = 0.048 and P = 0.01), and had a high level of IgA (P = 0.03) (Table 2). For other characteristics no differences between patients with or without MVK mutations were found.

Table 2.

Comparison of clinical features between patients with high level of IgD according to the presence of MVK mutations

 Mutations  
Characteristics Group A Patients with a high IgD serum level and MVK mutations n (%) Group B Patients with a high IgD serum level and no MVK mutation n (%) P-value (Group A vs Group B) 
Patients 19 19  
Gender (M) 14  
            (F) 11  
Starting factors 6 (31) 1 (5.2) 0.1 
Fever 18 (95) 18 (95) >0.99 
Abdominal pain 13 (68) 14 (74) >0.99 
Hepatomegaly 4 (21) 4 (21) 0.4 
Splenomegaly 6 (31) 6 (32) 0.5 
Diarrhoea 10 (52) 4 (21) 0.09 
Lymphadenopathy 14 (76) 5 (26) 0.01 
Cutaneous involvment 12 (63) 8 (42) 0.048 
Arthralgy 14 (73) 12 (63) 0.7 
Arthritis 5 (26) 3 (16) 0.7 
Pharingitis 3 (16) 4 (21) >0.99 
Aphtous ulcer 4 (21) 4 (21) 0.7 
Amyloidosis 0 (0) 0 (0) >0.99 
Neurological involvement 0 (0) 3 (16) 0.3 
Mean age of the first attack (year) 12.4 0.0006 
Mean duration of attacks (day) 5.3 6.08 0.11 
Frequency of attacks   0.4 
= 1attack/month 2 (10) 5 (30)  
> 1/month 16 (84) 12 (70)  
High level serum of IgA 12 (63) 4 (21) 0.03 
Normal level serum of IgA 2 (8.3) 6 (32)  
None carried out  
C-reactive protein>10 mg/l 18 (75) 15 (79) >0.99 
None carried out  
 Mutations  
Characteristics Group A Patients with a high IgD serum level and MVK mutations n (%) Group B Patients with a high IgD serum level and no MVK mutation n (%) P-value (Group A vs Group B) 
Patients 19 19  
Gender (M) 14  
            (F) 11  
Starting factors 6 (31) 1 (5.2) 0.1 
Fever 18 (95) 18 (95) >0.99 
Abdominal pain 13 (68) 14 (74) >0.99 
Hepatomegaly 4 (21) 4 (21) 0.4 
Splenomegaly 6 (31) 6 (32) 0.5 
Diarrhoea 10 (52) 4 (21) 0.09 
Lymphadenopathy 14 (76) 5 (26) 0.01 
Cutaneous involvment 12 (63) 8 (42) 0.048 
Arthralgy 14 (73) 12 (63) 0.7 
Arthritis 5 (26) 3 (16) 0.7 
Pharingitis 3 (16) 4 (21) >0.99 
Aphtous ulcer 4 (21) 4 (21) 0.7 
Amyloidosis 0 (0) 0 (0) >0.99 
Neurological involvement 0 (0) 3 (16) 0.3 
Mean age of the first attack (year) 12.4 0.0006 
Mean duration of attacks (day) 5.3 6.08 0.11 
Frequency of attacks   0.4 
= 1attack/month 2 (10) 5 (30)  
> 1/month 16 (84) 12 (70)  
High level serum of IgA 12 (63) 4 (21) 0.03 
Normal level serum of IgA 2 (8.3) 6 (32)  
None carried out  
C-reactive protein>10 mg/l 18 (75) 15 (79) >0.99 
None carried out  

Among patients without MVK mutations, the diagnosis of TRAPS was established with the detection of TNFRSF1A mutation in four patients. In 14 cases, no diagnosis was made (follow-up between 2 and 6 years), and no cases of amyloidosis were found.

In Table 3, we compared the characteristic of HIDS patients according to the presence of high IgD serum level. We did not find any difference between the two groups.

Table 3.

Characteristic features of HIDS patients

 Mutations  
Characteristics Group A1 HIDS with high level of serum IgD n (%) Group A2 HIDS with normal level of serum IgD n (%) P-Value (Group A1 vs Group A2) 
Patients 19  
Starting factors 6 (32) 1 (20) >0.99 
Ethnicity: Arab from Maghreb  
Caucasian 17 (90) 2 (40)  
Mediterranean origin  
Guadeloupe  
Fever 18 (95) 5 (100) >0.99 
Abdominal pain 13 (68) 4 (80) >0.99 
Hepatomegaly 4 (21) 2 (40) 0.58 
Splenomegaly 6 (32) 4 (80) 0.13 
Diarrhoea 10 (53) 3 (60) >0.99 
Lymphadenopathy 14 (74) 2 (40) 0.27 
Cutaneous involvment 12 (63) 3 (60) >0.99 
Arthralgy 14 (74) 2 (40) 0.27 
Arthritis 5 (26) 2 (40) >0.99 
Pharyngitis 3 (16) 1 (20) >0.99 
Aphtous ulcer 4 (21) 1 (20) >099 
Amyloidosis 0 (0) 1 (20) 0.23 
Mean age of the firs attack (year) 3.54 1.3 0.66 
Mean duration of attacks (day) 5.35 0.88 
High level serum of IgA 12 >0.99 
Normal level serum of IgA  
Non-related  
 Mutations  
Characteristics Group A1 HIDS with high level of serum IgD n (%) Group A2 HIDS with normal level of serum IgD n (%) P-Value (Group A1 vs Group A2) 
Patients 19  
Starting factors 6 (32) 1 (20) >0.99 
Ethnicity: Arab from Maghreb  
Caucasian 17 (90) 2 (40)  
Mediterranean origin  
Guadeloupe  
Fever 18 (95) 5 (100) >0.99 
Abdominal pain 13 (68) 4 (80) >0.99 
Hepatomegaly 4 (21) 2 (40) 0.58 
Splenomegaly 6 (32) 4 (80) 0.13 
Diarrhoea 10 (53) 3 (60) >0.99 
Lymphadenopathy 14 (74) 2 (40) 0.27 
Cutaneous involvment 12 (63) 3 (60) >0.99 
Arthralgy 14 (74) 2 (40) 0.27 
Arthritis 5 (26) 2 (40) >0.99 
Pharyngitis 3 (16) 1 (20) >0.99 
Aphtous ulcer 4 (21) 1 (20) >099 
Amyloidosis 0 (0) 1 (20) 0.23 
Mean age of the firs attack (year) 3.54 1.3 0.66 
Mean duration of attacks (day) 5.35 0.88 
High level serum of IgA 12 >0.99 
Normal level serum of IgA  
Non-related  

Discussion

As commonly performed in the past, the measurement of IgD levels in patients with symptoms of HIDS is still often used as a marker supporting the diagnosis of this autoinflammatory disorder. However, the discovery of MVK as the gene responsible for this disease led to a new classification between patients with or without mevalonate kinase deficiency. Although mutated patients were still labelled ‘HIDS’, the hyper IgD findings in these patients became puzzling and controversial. Indeed, some patients with periodic fever do not have increased IgD levels even they have mutations in MVK [7, 9, 10]. This enabled us to evaluate the usefulness of serum IgD level in the diagnosis of HIDS in a series of patients with a phenotype of recurrent fever.

In our series of 50 patients with a clinical suspicion of HIDS, the diagnosis was eventually established on the basis of a molecular diagnosis in 48% of patients. This reflects a high prevalence of the disease in a selected patient's population with a periodic fever syndrome, although there are so far no validated clinical criteria for the diagnosis of HIDS as they exist for FMF.

A high level of IgD was found in 19 HIDS patients, as well as in 19 patients who had a clinical suspicion of HIDS without MVK mutations. This high false-positive rate reveals that a patient with recurrent fever and elevated IgD level is only at risk of 50% to display a mevalonate kinase deficiency and confirms that a high serum IgD level is not specific for the diagnosis of HIDS (specificity = 0.27) [14]. Also, we analysed the two groups of patients with a high serum of IgD, in order to establish any distinctive features apart from mutation in MVK. Patients with MVK mutations had phenotypic similarities (Table 2), but pertinent clinical variables were found between the two groups. Age of onset at the first attack was significantly higher in patients without MVK mutations, and tended to have longer attacks (6.1 day vs 5.3 days for patients with MVK mutations) but without significant difference. These findings confirm the results already published by Simon et al. [14].

In our study, IgD levels were normal in five patients with MVK mutations. These patients had the same clinical presentations than MKD patients with high levels of IgD. The negative predictive value of a normal IgD level with mevalonate kinase deficiency is thus low in our series (58%).

Although this negative predictive value could be particularly low and biased in part by the low number of patients with a normal IgD level and no MVK mutation, we think that in case of clinical suspicion of HIDS a search for MVK mutations could be done whatever the level of serum IgD.

Few cases of HIDS with normal IgD serum levels have been described previously [9, 10].

In our study, these patients were respectively 8-months, 16-months, 2-years, 3-years and 4-years–old. We know that in normal children of <1 year and particularly in neonates, the rate of IgD is lower than in adults [15].

In fact, in the whole population of 144 patients tested for MVK mutations, only 24 have two mutations. Out of 144, 94 have been excluded from the study, as IgD's were not available. This bias, however, restricts the strength of predictive values in our study but does not strongly affect other indices.

Globally, the clinical relevance of the IgD measurement for the diagnosis of HIDS in our population appears poor, as reflected by likelihood ratios which are both close to 1.

The mechanism of the increased value of serum IgD in HIDS is still unexplained.

Simon et al. [16] have shown in a series of 22 Dutch adult patients with MKD that a high IgD level is not a consequence of the inflammatory phenotype, and the serum IgD concentration may vary greatly during life without correlation with clinical symptoms or frequency of attacks. Also, no difference in IgD serum concentrations was found in samples taken during remission or during fever attacks (median 639 UI/ml vs 603 UI/ml) and no correlation between IgD and CRP values.

Taken together, these two studies lead to the conclusion that the high IgD level is an epiphenomenon rather than to consider it as a central to the pathogenesis of HIDS. This point is coherent with the observation that an elevated serum IgD is not a specific marker for HIDS as it has been previously reported.

Also, high levels of serum IgD were reported to be increased in other hereditary auto-inflammatory syndrome (FMF, TRAPS), malignancies and infectious diseases [17–20]. In the 19 patients of our cohort with high level of IgD without MVK mutations, four patients had a final diagnosis of TRAPS including two patients in the same family. In these four patients, the diagnosis of TRAPS could have been considered first because of the nature of clinical symptoms and of the pedigree suggesting a dominant mode of inheritance. This emphasizes the relevance of clinical data to distinguish between TRAPS and HIDS. Despite a follow-up of 2–6 years, no diagnosis was done in these 14 cases for which the actual diagnosis remains thus to be established. All these 14 cases appear as sporadic, without a family history, but this cannot eliminate completely a possible genetic cause. It should be kept in mind that among auto-inflammatory syndromes, the CINCA phenotype had been considered for a long time as a sporadic disease, until a dominant mode of inheritance could be observed in some pedigrees [21].

However, a number of these patients could be also phenocopies of MKD. Further studies are needed to decipher the mechanisms and aetiology of the diseases in this group of patients with unexplained recurrent fever syndromes.

We would like to propose to use the term of MKD for patients who have a biochemical deficiency of mevalonate kinase and mutations in the MK gene, whatever their level of serum IgD. In fact, it looks uneasy, from a semantic viewpoint, to name ‘HIDS’ patients with normal IgD level and a MK deficiency. The term HIDS which contains the word syndrome and thus does not define a disease could be still used for the group of patients who have a recurrent fever syndrome and an elevated IgD concentration.

graphic

Acknowledgements

Urinary mevalonic acid analysis was performed by Christine Vianet-Saban and Daniel Rabier.This work was supported by a grant from PHRC AOM97201 and by a grant from the GIS Maladies Rares.

The authors have declared no conflicts of interest.

References

1
Van der Meer
JW
Vossen
JM
Radl
J
, et al.  . 
Hyperimmunoglobulinemia D and fever: a new syndrome
Lancet
 , 
1984
, vol. 
1
 (pg. 
1087
-
90
)
2
Drenth
JPH
Haagsma
CJ
Van
der Meer
JW and the International Hyper-IgD Study Group.
Hyperimmunoglobulinemia D and periodic fever syndrome. The clinical spectrum of a series of 50 patients
Medecine
 , 
1994
, vol. 
73
 (pg. 
133
-
44
)
3
Hiemstra
I
Vossen
JM
Van der Meer
JW
Weemaes
CM
Out
TA
Zegers
BJ
Clinical and immunological studies in patients with increased serum IgD level
J Clin Immunol
 , 
1989
, vol. 
9
 (pg. 
393
-
400
)
4
Lachmann
HJ
Goodman
HJ
Andrews
PA
, et al.  . 
AA amyloidosis complicating hyperimmunoglobulinemia D with periodic fever syndrome: a report of two cases
Arthritis Rheum
 , 
2006
, vol. 
54
 (pg. 
2010
-
14
)
5
Obici
L
Manno
C
Onetti Muda
A
, et al.  . 
First report of systemic reactive (AA) amylodosis in a patient with the hyperimmunoglobulinemia D with periodic fever syndrome
Arthritis Rheum
 , 
2004
, vol. 
9
 (pg. 
2966
-
69
)
6
Drenth
JPH
Cuisset
L
Grateau
G
, et al.  . 
Mutations in the gene encoding mevalonate kinase cause hyper-IgD and periodic fever syndrome
Nat Genet
 , 
1999
, vol. 
22
 (pg. 
178
-
81
)
7
Houten
SM
Kuis
W
Duran
M
, et al.  . 
Mutations in MVK, encoding mevalonate kinase, cause hyperimmunoglobuliemia D and periodic fever syndrome
Nat Genet
 , 
1999
, vol. 
22
 (pg. 
175
-
77
)
8
Cuisset
L
Drenth
JP
Simon
A
, et al.  . 
Molecular analysis of MVK mutations and enzymatic activity in hyper-IgD and periodic fever syndrome
Eur J Hum Genet
 , 
2001
, vol. 
9
 (pg. 
260
-
66
)
9
Di Rocco
M
Caruso
U
Waterham
HR
Picco
P
Loy
A
Wanders
RJA
Mevalonate kinase deficiency in a child with periodic fever and without hyperglobulinaemia D
J Inherit Met Dis
 , 
2001
, vol. 
24
 (pg. 
411
-
12
)
10
Franck
TS
Hyperimmunoglobulinemia D and periodic fever syndrome (HIDS) in a child with normal serum IgD, but increased serum IgA concentration
J Pediatr
 , 
2003
, vol. 
143
 (pg. 
127
-
29
)
11
Gibson
KM
Lohr
JL
Broock
RL
, et al.  . 
Mevalonate kinase in lysates of cultured human fibroblasts and lymphoblasts: kinetic properties, assay conditions, carrier detection and measurement of residual activity in a patient with mevalonic aciduria
Enzyme
 , 
1989
, vol. 
41
 (pg. 
47
-
55
)
12
Hoffmann
GF
Brendel
SU
Sharfschwerdt
SR
Shin
YS
Speidel
IM
Gibson
KM
Mevalonate kinase assay using DEAE-cellulose column chromatography for first-trimester prenatal diagnosis and complementation analysis in mevalonic aciduria
J Inherit Metab Dis
 , 
1992
, vol. 
15
 (pg. 
738
-
46
)
13
Lindenthal
B
von Bergmann
K
Determination of urinary mevalonic acid using isotope dilution technique
Biol Mas Spectrom
 , 
1994
, vol. 
23
 (pg. 
445
-
50
)
14
Simon
A
Cuisset
L
Vincent
MF
, et al.  . 
Molecular analysis of the MVK gene in a cohort of patients with the hyper-immunoglobuline D and periodic syndrome: its application as a diagnostic tool
Ann Intern Med
 , 
2001
, vol. 
135
 (pg. 
338
-
43
)
15
Levan-Petit
I
Cardonna
J
Garcia
M
, et al.  . 
Sensitive ELISA for human immunoglobulin D measurement in neonate, infant and adult sera
Clin Chem
 , 
2000
, vol. 
46
 (pg. 
876
-
8
)
16
Simon
A
Bijzet
J
Voorbij
HAM
Martovani
A
Van der Meer
JW
Drenth
JP
Effect of inflammatory attacks in the classical type hyper-IgD syndrome on immunoglobulin D, cholesterol and parameters of the acute phase response
J Intern Med
 , 
2004
, vol. 
256
 (pg. 
247
-
53
)
17
Blade
J
Lust
JA
Kyle
RA
Immnunoglobulin D multiple myeloma: presenting features, response to therapy, and survival in a series of 53 cases
J Clin Oncol
 , 
1994
, vol. 
12
 (pg. 
2398
-
404
)
18
Raiteri
R
Albonico
M
Deiana
R
Marietti
G
Sinico
A
Serum IgD behaviour in HIV 1 infected patients
Allergol Immunopathol
 , 
1991
, vol. 
19
 (pg. 
6
-
10
)
19
Walzer
PD
Amstroug
D
Weissman
P
Tan
C
Serum immunoglobulin levels in chilhood Hodgkin's disease. Effect of splenectomy and long term follow-up
Cancer
 , 
1980
, vol. 
45
 (pg. 
2084
-
89
)
20
Schall
PR
Periodic fever syndromes
Curr Opin Pediatr
 , 
2000
, vol. 
12
 (pg. 
563
-
66
)
21
Feldmann
J
Prieur
AM
Quartier
P
, et al.  . 
Chronic infantile neurological cutaneous and articular syndrome is caused by mutations in CIAS1, a gene highly expressed in polymorphonuclear cells and chondrocytes
Am J Hum Genet
 , 
2002
, vol. 
71
 (pg. 
198
-
203
)

Comments

0 Comments