Persisting Mixed Cryoglobulinemia in Chikungunya
Manuela Oliver1, Marc Grandadam2, Catherine Marimoutou3, Christophe Rogier3, Elisabeth
Botelho-Nevers4, Hugues Tolou2, Jean-Luc Moalic1, Philippe Kraemer4, Marc Morillon5, Jean-Jacques
Morand6, Pierre Jeandel3, Philippe Parola7, Fabrice Simon5*
1Laboratoire de Biochimie, Ho ˆpital d’Instruction des Arme ´es Laveran, Marseille, France, 2Unite ´ de Virologie Tropicale, Institut de Me ´decine Tropicale du Service de Sante ´
des Arme ´es, Marseille, France, 3Comite ´ Recherche, HIA Laveran, Marseille, France, 4Service de Pathologie Infectieuse et Tropicale, Ho ˆpital d’Instruction des Arme ´es
Laveran, Marseille, France, 5Laboratoire de Biologie, HIA Laveran, Marseille, France, 6Service de Dermatologie, HIA Laveran, Marseille, France, 7Service des Maladies
Infectieuses et Tropicales, Centre Hospitalo-Universitaire Nord, Assistance Publique Ho ˆpitaux de Marseille, Marseille, France
Background: Chikungunya virus (CHIKV), an arbovirus, is responsible for a two-stage disabling disease, consisting of an
acute febrile polyarthritis for the first 10 days, frequently followed by chronic rheumatisms, sometimes lasting for years. Up
to now, the pathophysiology of the chronic stage has been elusive. Considering the existence of occasional peripheral
vascular disorders and some unexpected seronegativity during the chronic stage of the disease, we hypothesized the role of
Methods: From April 2005 to May 2007, all travelers with suspected CHIKV infection were prospectively recorded in our
hospital department. Demographic, clinical and laboratory findings (anti-CHIKV IgM and IgG, cryoglobulin) were registered
at the first consultation or hospitalization and during follow-up.
Results: Among the 66 travelers with clinical suspicion of CHIKV infection, 51 presented anti-CHIKV IgM. There were 45
positive with the serological assay tested at room temperature, and six more, which first tested negative when sera were
kept at 4uC until analysis, became positive after a 2-hour incubation of the sera at 37uC. Forty-eight of the 51 CHIKV-
seropositive patients were screened for cryoglobulinemia; 94% were positive at least once during their follow-up. Over 90%
of the CHIKV-infected patients had concomitant arthralgias and cryoglobulinemia. Cryoglobulin prevalence and level drop
with time as patients recover, spontaneously or after short-term corticotherapy. In some patients cryoglobulins remained
positive after 1 year.
Conclusion: Prevalence of mixed cryoglobulinemia was high in CHIKV-infected travelers with long-lasting symptoms. No
significant association between cryoglobulinemia and clinical manifestations could be evidenced. The exact prognostic
value of cryoglobulin levels has yet to be determined. Responsibility of cryoglobulinemia was suspected in unexpected false
negativity of serological assays at room temperature, leading us to recommend performing serology on pre-warmed sera.
Citation: Oliver M, Grandadam M, Marimoutou C, Rogier C, Botelho-Nevers E, et al. (2009) Persisting Mixed Cryoglobulinemia in Chikungunya Infection. PLoS
Negl Trop Dis 3(2): e374. doi:10.1371/journal.pntd.0000374
Editor: A. Desiree La Beaud, Case Western Reserve University School of Medicine, United States of America
Received May 6, 2008; Accepted January 6, 2009; Published February 3, 2009
Copyright: ? 2009 Oliver et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the French Service de Sante ´ des Arme ´es. The funders had no role in study design, data collection and analysis, decision to
publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: email@example.com
Chikungunya fever is an emerging arboviral disease character-
ized by a brief fever, headache and myalgias, occasional
evanescent rash, inflammatory polyarthralgias, arthritides or
tenosynovitis that can last for months to years [1–4]. Chikungunya
virus (CHIKV) was identified in the 1950s in Africa , and soon
after in Asia . It can be responsible for major epidemics,
sometimes separated by silent periods . From 2004 to 2006, a
giant CHIKV outbreak successively swept out Kenya and most
islands of western Indian Ocean . In Re ´union Island, the
outbreak was explosive at the beginning of 2006 with a pick of
45,000 cases per week. Up to 2006 June 1st, about one third of the
770,000 residents had been infected . Another huge outbreak
recently stroke India with 2 to 7 million estimated cases , and is
currently spreading to Southeastern Asia . During this period,
Chikungunya fever was also identified in more than 1,000 travelers
returning from the epidemic areas to European countries [1,11,12]
and the USA [13,14]. CHIKV-infected travelers included viremic
patients who returned home to countries where competent vectors
are present, raising serious concern for the globalization of the
disease [7,15]. The Italian outbreak in August 2007 has
demonstrated the reality of this threat .
During the recent CHIKV outbreaks, previously described
clinical features [3,4] as well as the low rate of asymptomatic
infections were confirmed [7,17]. The disease was also responsible
www.plosntds.org1February 2009 | Volume 3 | Issue 2 | e374
for unusual and unfrequent complications, including severe
newborn infections after peripartum mother-to-infant transmis-
sion, meningo-encephalitis, hepatitis, myocarditis, severe epider-
molysis [17,18] and lead to surmortality . Transitory peripheral
vascular disorders (PVD), mostly Raynaud syndrome, were also
observed few weeks after the disease onset . Considering
persistent arthralgias and occasional PVD, we hypothesized that
cryoglobulin could be involved in the pathophysiology of the
disease, as described for hepatitis C infection. In this infection high
level of mixed cryoglobulinemia is commonly detected in patients
with chronic viral replication and is strongly associated with liver
damage and peripheral neuropathies [19,20].
From April 2005 throughout May 2007, all patients with
possible imported CHIKV infection (recent travel in Indian
Ocean islands and presence or history of fever and/or arthralgias)
were prospectively recorded at Laveran Military Hospital in
Marseille, France. The criteria for confirmed cases were i)
presence of specific anti-CHIKV IgM and/or positive RT-PCR
and/or isolation of CHIKV from blood and/or specific anti-
CHIKV IgG, ii) recent clinical feature consistent with CHIKV
infection, iii) no other etiology identified. Demographic, clinical
and laboratory findings were registered for all patients at their first
consultation or hospitalization and during follow-up. For patients
seen more than 10 days after the onset of illness, early clinical
features were identified using a retrospective questionnaire. The
clinical status of each patient was actively monitored by the same
physician (FS) during consultations and/or by phone calls every 2–
3 months. The early stage is defined as the first 10 days of clinical
disease while second stage is defined as symptoms and signs
persisting more than 10 days after disease onset .
The same physician (FS) orally informed all patients about the
requirement of blood samples to diagnose the aetiology of their
polyarthralgia. The script for obtaining oral consent was accepted
by the Institutional Review Board at Laveran Hospital. All
patients gave their oral consent, which was noted in their
individual medical file. A single blood sample of less than 50 cc
was taken after consultation, when clinically required; no
supplementary blood samples were taken for research.
At each consultation, patients were tested for CHIKV serology
using in-house IgM-capture and IgG-sandwich enzyme-linked
immunosorbent assays . Serologies were performed first on sera
kept at 4uC before analysis and second, on sera kept at 37uC until
analysis. IgG and IgM for West Nile, dengue and Rift Valley fever
viruses were also assayed. On acute sera, RT-PCR and isolation of
CHIKV in Vero-E6 cells were attempted . Cryoglobulinemia
was screened using the following procedure: i) collection of 14 ml
of blood in pre-warmed tubes at the hospital’s biochemistry
laboratory, ii) clotting at 37uC for 3 hours, iii) centrifugation at
37uC (3,000 RPM for 10 minutes), iiii) freezing of the pellet at 4uC
for 10 days, to allow generation of cryoprecipitates. Purification,
characterization and quantification of cryoglobulins were per-
formed by the same operator (MO), as previously described .
Cryoglobulins were classified as type I for monoclonal component,
type II for one monoclonal component associated with polyclonal
immunoglobulins, type II–III for more than one monoclonal
component associated with polyclonal immunoglobulins, and type
III for polyclonal immunoglobulins [23,24]. Type III cryoglobu-
linemia was considered positive when level was higher than 5 mg/
L , whereas type II and type II–III cryoglobulins were
considered positive regardless of concentration level . CHIKV
RNA in cryoprecipitates was searched using RT-PCR, as
described elsewhere .
Patients also underwent immunological tests: (i) determination
of C3 and C4 complement components using an immunonephele-
metric method (Dade Behring Paris France), (ii) determination of
total haemolytic complement using the total haemolytic comple-
ment kit (The Binding Site Saint Egreve France), (ii) search for
rheumatoid factors, by latex immunoagglutination (Biome ´rieux
Marcy l’Etoile France), (iv) detection of antinuclear antibodies by
indirect immunofluorescence (Eurobio Paris France). Serological
screening for hepatitis C virus (HCV) infection was systematically
performed using sandwich enzyme-linked immunosorbent assays
(Biorad Marne La Coquette France). After 6 months, blood testing
for asymptomatic patients was restricted to CHIKV serology and
detection of cryoglobulinemia.
Chi 2, Kruskall-Wallis and exact Fisher’s, two-tailed probability
tests were used with a significant p value of 0.05 (Stata 9.0
Software, StataCorp College Station, Texas, USA).
Patients and First Screening for CHIKV Infection
During the 25 month-study, 66 French patients with clinical
suspicion of CHIKV infection were prospectively included.
Among them, 51 presented with anti-CHIKV IgM (see below).
Fifteen patients remained seronegative for both anti-CHIKV IgM
and IgG. No patients had detectable CHIK virus in sera (RT PCR
and CHIK V isolation on Vero cell were negative).
Sex ratio (M/F) and median age in seropositive patients were
respectively 1.04 and 54 years (range: 21 y–78 y), versus 0.67 and
47 years (21 y–75 y) in seronegative patients (no statistical
difference). Six CHIKV-infected patients were lost to follow-up
after the first consultation. The 45 other patients have been
followed in our medical unit for a median duration of 14 months
Chikungunya virus is present in tropical Africa and Asia
and is transmitted by mosquito bites. The disease is
characterized by fever, headache, severe joint pain and
transient skin rash for about a week. Most patients
experience persisting joint pain and/or stiffness for months
to years. In routine practice, diagnosis is based upon
serology. Since 2004 there has been an ongoing giant
outbreak of Chikungunya fever in East Africa, the Indian
Ocean Islands, India and East Asia. In parallel, more than
1,000 travelers were diagnosed with imported Chikungu-
nya infection in most developed countries. Considering
the clinical features of our patients (joint pain), we
hypothesized that cryoglobulins could be involved in the
pathophysiology of the disease as observed in chronic
hepatitis C infection. Cryoglobulins, which are immuno-
globulins that precipitate when temperature is below
37uC, can induce rheumatic and vascular disorders. From
April 2005 through May 2007, we screened all patients
with possible imported Chikungunya infection for cryo-
globulins. They were present in over 90% of patients, and
possibly responsible for the unexpected false negativity of
serological assays. Cryoglobulin frequency and levels
decreased with time in recovering patients.
Cryoglobulinemia in Chikungunya Infection
www.plosntds.org2 February 2009 | Volume 3 | Issue 2 | e374
(range: 54 days–25 months), through 2 to 6 consultations per
patient; resulting in a total of 138 consultations for CHIKV
patients performed up to May 2007.
Table 1 summarizes the demographic, epidemiological and
clinical events of the 51 confirmed cases. Ninety-eight percent of
the seropositive cases suffered at least once with arthralgia, 71%
with tenosynovitis and 20% with transitory PVD. Thirteen
patients experienced at least one clinical relapse during follow-
up, i.e. became symptomatic again after at least one symptom free
month, mostly with subacute arthralgias in hands and feet.
Thirteen patients developed de novo transitory PVD, mainly in
fingers and sometimes in toes, within the second and third months
after the disease onset. No other aetiology (drug, auto-immune
disorder, Buerger disease, local trauma) than CHIKV was
identified for relapses or PVD. No morphological changes were
observed in the 3 patients for whom a digital capillaroscopy was
performed. PVD was not significantly associated with gender.
Conversely, tenosynovitis was significantly more frequent in
women, both on the day of consultation (p=0.01) or during the
previous month (p=0.04).
Table 1. Characteristics of 51 patients with Chikungunya
infection imported from Indian Ocean Islands between April
2005 and October 2006, to Laveran Military hospital, Marseille,
Demographic and epidemiological data
Median age (ranges)54.0 years (21–78 y)
Sex ratio M/F 1.04 (26/25)
Chronic general or rheumatic disease 37
Year of contamination
Location of contamination
Reunion or Mauritius1
Location of disease onset
Western Indian Ocean34
Mean duration of stay
Residents in Western Indian Ocean9 days
No residents54 days
During early (acute) stage4
During chronic stage8
Acute stage (within the first 10 days)
Fever (mean duration)47 (3.5 days)
Gingival or nasal bleeding3
Peripheral arthralgias/ arthritis50
Number of involved joint groups 51
Less than 1020
Ten or more30
Hands and/or feet involvement48
Symmetry of joint involvement44
Nervous tunnel syndrome3
Peripheral vascular disorder3 (finger coldness)
Complication 1 (acute myocarditis)
Chronic stage (after the 10thday)
Chronic joint pain and/or stiffness48
Hand and/or feet involvement47
Symmetry of joint involvement 43
Exacerbation of pre-existing pains14
Nervous tunnel syndrome17
Peripheral vascular disorder
Short-term general corticotherapy26
This table reports the most frequent signs and symptoms, regardless of
Clinical characteristics of 39 of these patients have been described elsewhere
Table 1. Cont.
Figure 1. Left side: Red stage of Raynaud syndrome and
swelling of finger joints in a 36-year-old CHIKV-infected
woman 6 weeks after disease onset; right side: concomitant
cryoprecipitate in her blood.
Cryoglobulinemia in Chikungunya Infection
www.plosntds.org3February 2009 | Volume 3 | Issue 2 | e374
Among the 51 CHIKV-infected patients, cryoglobulinemia was
screened for 48, i.e. 42 patients on first consultation and 6 later
during follow-up. Cryoglobulinemia screening was repeated 2 to 5
times for 38 patients during the study period. The median delay
between CHIKV acute symptomatology onset and first cryoglob-
ulinemia testing was 45 days (range: 5–640 d). On first
consultation, 37/42 patients (88%) were cryoglobulinemic. Among
the five patients without cryoglobulinemia: i) three symptomatic
patients respectively free of cryoglobulinemia on days 5, 16 and 36
after disease onset, developed cryoglobulinemia on days 27, 45
and 69, respectively; ii) one symptomatic patient free of
cryoglobulinemia on his first test at day 124 (a week after a
short-term corticotherapy) became positive with a type II on day
204; iii) one patient had been symptom-free for weeks and was not
cryoglobulinemic when tested on day 107 after disease onset.
Among the 6 patients first tested at the second consultation, 4
were cryoglobulinemic and two remained negative for cryoglob-
ulin on day 143 and day 355, respectively, while they were both
symptom-free. Finally, 94% of the 48 CHIKV-infected tested
patients were positive for cryoglobulinemia at least once during
During the study period, 118 cryoglobulin assays were
performed, 83 were positive. There were 61% (51/83) type II,
30% (25/83) type II–III and 8% (7/83) type III. Monoclonal IgM-
k and IgM-l components were found in 56/83 cryoglobulins
(68%) and 48/83 (58%), respectively; polyclonal IgM components
in 83/83 cryoglobulins (100%). Monoclonal IgA was found in only
2/83 cryoglobulins. Thus, all cryoglobulins associated with
CHIKV infection were mixed cryoglobulins.
The cryoglobulin concentration could be determined for 79/83
cryoglobulinemias. The median cryoglobulin level was 8 mg.L21
Table 2. Cryoglobulin types and levels regarding delay after Chikungunya infection onset, at Laveran Military hospital, Marseille,
Type IIaPrevalence (%)
Type II–IIIbPrevalence (%)
Type III Prevalence (%)
All typescPrevalence (%)
0–30 days2/16 (12.5%) 10/16 (62.5%) [14.5 mg.L21] 2/16 (12.5%) [2.7 mg.L21] 2/16 (12.5%) [6.8 mg.L21] 14/16 (87.5%) [7.7 mg.L21]
31–90 days3/28 (10.7%)14/28 (50.0%) [10.1 mg.L21] 7/28 (25.0%) [25.2 mg.L21] 4/28 (14.3%) [35.9 mg.L21] 25/28 (89.3%) [10.4 mg.L21]
91–180 days 10/27 (37.0%)10/27 (37.0%) [6.54 mg.L21] 7/27 (25.9%) [26.2 mg.L21] 0/27 (0%)17/27 (63.0%) [11.7 mg.L21]
181–747 days 20/47 (42.6%) 17/47 (36.2%) [3.1 mg.L21] 9/47 (19.1%) [4.1 mg.L21] 1/47 (2.1%) [21.6 mg.L21] 27/47 (57.4%) [3.7 mg.L21]
cp=0.002 (Kruskall-Wallis test for distribution between the different delays).
Figure 2. Evolution of all types cryoglobulin levels over time in 51 Chikungunya-infected travelers, Laveran Military Hospital,
Marseille, France. The sample results are presented as follows: box represents the interquartile range, the included line figuring the median; lines
out of the box represent range with extreme outliers figured as points.
Cryoglobulinemia in Chikungunya Infection
www.plosntds.org4 February 2009 | Volume 3 | Issue 2 | e374
(range: 3–165). No CHIKV was detected by RT-PCR in 24 tested
Cryoglobulinemia and serological assay for CHIKV
When performed on sera kept at 4uC before
analysis, detection of anti-CHIKV IgM was positive for 45 /66
patients. Among them, cryoglobulinemia was found in 40 patients
(89%). Moreover, cryoglobulins were detected in 11 (52%) of the
21 initially seronegative patients. For these 11 patients new ELISA
were conducted on sera kept at 37uC until analysis, to allow
dissolution of putative cryoprecipitates. Anti-CHIKV IgM were
found in 6 of these patients and anti CHIKV IgG in 4. Pre-
warmed sera were characterized by high levels of specific
antibodies and presence of cryoglobulin in 5 /6 cases. When
summarizing, a total of 51 patients were serologically confirmed
Table 3. Clinical and biological status of 51 patients with Chikungunya infection at successive follow-up delays, Laveran Military
hospital, Marseille, France (M: month).
M1 M2M3 M5M7 M10M13
Patients with available status51 50 474746 4431
Symptomatic patients50 47 43 3832 2818
Cryoglobulemic/all tested patients13/1516/17 11/13 10/1511/189/1310/19a
Cryoglobulinemic patients/symptomatic tested patients12/14 16/1710/12 8/118/13 5/98/13
Cryoglobulinemic patients/asymptomatic tested patients1/1- 1/12/43/6 4/42/6
Patients with anti-CHIKV IgM /all tested patients 18/18 17/1712/12 12/1512/16 5/12 2/18
Patients with both anti-CHIKV IgM & cryoglobulinemia/tested patients13/15 16/1711/12 9/15 7/16 5/122/18b
Patients with anti-CHIKV IgG/tested patients12/1816/17 11/12 14/1514/1610/1213/18
aone patient underwent two cryoglobulin tests, the one performed one week after general corticotherapy is not presented here.
bthese two patients remained symptomatic even after 15 months.
Figure 3. Individual follow up of a 62-year-old female patient with CHIKV infection. X-axis, month of follow-up; Y-axis, Cryoglobulin level
Cryoglobulinemia in Chikungunya Infection
www.plosntds.org5 February 2009 | Volume 3 | Issue 2 | e374
with CHIKVinfection. Ninety-fourpercent where cryoglobulinemic
at least once during follow up, while only 27% (4/15) of CHIKV-
seronegative patients were cryoglobulinemic (p,0.0001). All 51
patients but one (98%) were positive for anti-CHIKV IgM and 88%
for IgG. The remaining patient, unless presenting with typical
CHIKV-clinical symptoms, became anti-CHIKV IgG positive only
when tested 124 days after disease onset.
CHIKV-associated cryoglobulinemia, clinical features
between age and cryoglobulin levels (p=0.012). As arthralgias
and cryoglobulinemia were quasi constant (prevalence 98% and
94%, respectively) no statistical association could be tested. No
concentration according to presence of tenosynovitis or number
of joints involved (less or more than 10). Although patients with
PVD presented with more type II–III cryoglobulins (25%) than
patients without PVD (13%), the difference was not significant.
Nevertheless a cryoglobulin concentration higher than 8 mg.L21
(median concentration) was present in 7/31 samples from patients
suffering with PVD versus 2/39 samples from patients without
PVD (p=0.06). Two specific clinical cases are particularly
illustrative for pathophysiology: (i) the type II–III cryoglobulin
level of a 36-year-old woman rose from 32 mg.L21on day 40 to
104 mg.L21on day 60, while she was just developing a severe
Raynaud syndrome (Figure 1); (ii) a 20-year-old woman, whose
severe acute stage was marked by transitory myocarditis and finger
wasa negative correlation
coldness, also presented high levels of type III cryoglobulin, up to
165 mg.L21on day 85. No other specific clinical sign was
presented by patients whose cryoglobulin level was below
Twenty-six patients received short-term general corticotherapy,
mainly prescribed for disabling distal polyarthritis, multiple
tenosynovitis or severe PVD, when non-steroidal anti-inflamma-
tory drugs (NSAIDs) were contraindicated (4 cases) or ineffective
(22 cases). All of them experienced definitive or transitory
improvement of tendon pains, PVD and sometimes arthralgias
after this treatment.
The parallel evolution of symptoms and cryoglobulinemia
under corticotherapy in a 54-year-old CHIKV-infected woman
was particularly illustrative. On day 34 after disease onset, she was
disabled by severe polyarthritis and tenosynovitis and a type II
cryoglobulin was identified at 15.2 mg.L21. As NSAIDs were
contraindicated because of oral anticoagulation, she received
general corticotherapy with progressive withdrawal within the 3rd
month. At the end of this treatment, her clinical status improved
dramatically while cryoglobulinemia disappeared. After 8 months
of evolution, a low level (4.06 mg.L21) of type II–III cryoglobulin
was detected, while she only complained of mild stiffness in the
fingers. This case was suggestive for an unexpected role of MC in
pathophysiology of late CHIKV manifestations.
Cryoglobulin was screened in 39 patients throughout the
Figure 4. Individual follow up of a 41-year-old female patient with CHIKV infection. X-axis, month of follow-up; Y-axis, Cryoglobulin level
Cryoglobulinemia in Chikungunya Infection
www.plosntds.org6February 2009 | Volume 3 | Issue 2 | e374
follow-up period. The earliest cryoglobulins were detected on
day 6 after disease onset. Cryoglobulin prevalence rate was
negatively associated with elapsed time after disease onset
(p=0.002). Prevalence remained stable around 80% within the
first three months, and decreased to 57% in the second semester
(Table 2). Cryoglobulins were still detected in 55% of the 18
patients evaluating for more than a year. Types II and III
cryoglobulins seemed to be detected earlier than type II–III
cryoglobulins (median delay: 36.5 and 41 days versus 58.5 days,
respectively; p=0.07). The evolution of cryoglobulin levels over
time after CHIKV infection onset is presented in Figure 2. Type
II and II–III cryoglobulin levels significantly changed over time
(type II, p=0.008; type II–III, p=0.04; table 2). Interestingly, a
type II cryoglobulin at 6 mg.L21was detected in a Comorian
woman who was still painful one year after her infection in the
Comoros archipelago. On day 596, no cryoglobulin was
detected, whereas the patient had become asymptomatic for 4
Table 3 summarizes the main clinical and biological data of the
CHIKV-infected patients at successive times during follow-up.
Figures 3, 4, 5, 6 and 7 illustrate examples of clinical and biological
evolution after CHIKV infection.
Other Laboratory Testing
Neither antinuclear antibody nor rheumatoid factors were
found during follow-up. C3, C4 complement fractions and
haemolytic complement levels remained normal. All patients were
seronegative for HCV.
Our study was an empiric prospective study of symptomatic
CHIKV-infected travelers coming back from Western Indian
Ocean. Thus, follow-up for cryoglobulinemia was not performed
in seronegative patients in the absence of ethic committee consent.
Up to now, pathogenesis of CHIKV-induced rheumatism
remains unknown. Fourie et al. identified low titers of rheumatoid
factor in CHIKV infection , whereas no antinuclear antibodies
or rheumatoid factors were detected here like in previous studies
. We identified for the first time the association between
CHIKV infection and cryoglobulinemia among infected travelers.
As no type I cryoglobulin (composed of one monoclonal
immunoglobulin) was detected in our patients, CHIKV seems to
induce only mixed cryoglobulinemia (MC), either type II, II–III or
MC has already been described in miscellaneous acute and
chronic infections [19,20]. In these cryoglobulinemia seemed to
rapidly decrease and disappear within a few weeks of pathogen
clearance, although persisting MC has been observed in few
chronic infections, the most common being chronic hepatitis C
virus (HCV) (MC prevalence range: 20–55%) [19,20]. In our
study, the strength of association between CHIKV infection and
MC was much higher 94%. A biased high prevalence of CHIKV-
Figure 5. Individual follow up of a 36-year-old female patient with CHIKV infection. X-axis, month of follow-up; Y-axis, Cryoglobulin level
Cryoglobulinemia in Chikungunya Infection
www.plosntds.org7 February 2009 | Volume 3 | Issue 2 | e374
MC due to use of a highly sensitive method can be ruled in regard
of the frequent negativity of cryoglobulinemia detection over time
The persistence of CHIKV-associated MC (CHIKV-MC) was
unexpected, as the CHIKV genome has never been detected in
blood after 12 days of disease evolution, even in chronic
symptomatic patients . Consistently, we failed to detect
CHIKV-RNA in the cryoprecipitates, as described in chronic
HCV infection [26,27]. The absence of chronic viremia could also
explain the much lower level of CHIKV-MC than those observed
in chronic HCV infection . Ozden et al. recently showed the
presence of CHIKV in muscle satellite cells of a non-immuno-
compromised patient who was still painful three months after
disease onset . Moreover, Jaffar-Bandjee et al., using molecular
tools, evidenced persistence of CHIKV in perivascular macro-
phages of synovial tissue in a chronic elbow hygroma of a patient
infected for one year . This discovery suggests that CHIKV
can induce a chronic infection with active replication, conversely
to most other arboviruses. Persistence of CHIKV also seemed to
stimulate host immunity, as inflammation with macrophages and
T cells was locally concomitantly observed. Thus, viral replication
could be involved in the prolonged persistence of anti-CHIKV
IgM and, as a consequence, of CHIKV-MC in patients with long-
persisting symptoms. CHIKV-MC should be the possible ‘‘missing
link’’ between CHIKV and some symptoms of the chronic stage.
The second main finding is the existence of false seronegativity
in one out of three and a half CHIKV-infected patient with typical
clinical presentation of CHIKV infection when using ‘‘classical’’
ELISA assay at room temperature. This phenomenon can be
responsible for non-recognition of CHIKV infection in individuals
with chronic rheumatism, leading to useless explorations, as well as
underestimation of seroprevalence in an endemic/epidemic area.
Considering the high prevalence of CHIKV-MC in our cohort, we
assume that cryoglobulinemia may be a significant factor in
misdiagnosing the disease when using ‘‘conventional’’ serology.
Specific anti-CHIKV antibodies, i.e. IgM and IgG, could all be
trapped in the cryoprecipitate, as already described in chronic
HCV infection . Therefore, when facing a patient with a
clinical suspicion of CHIKV infection and with paradoxical
seronegativity, we recommend the following procedure: at least
pre-warmed the serum at 37uC before serology testing or,
preferably, manage the blood sample as required for any
cryoglobulin research: sampling and centrifugation at 37uC,
decantation and serum pre-warming before the ELISA assays.
However, considering the high prevalence of MC among our
CHIKV-seronegative patients (40%) and the unexpected persis-
tent seronegativity in a few patients with a high clinical suspicion
of CHIKV-infection, we cannot exclude some residual false
negative results due to the precipitation of CHIKV-MC before
Chronic MC is commonly associated with various symptoms,
mostly arthralgias, purpura, weakness and Raynaud syndrome
[19,20]. In our study, no CHIKV-infected patient presented the
complete clinical triad associated with MC, but arthralgias were
Figure 6. Individual follow up of a 62-year-old male patient with CHIKV infection. X-axis, month of follow-up; Y-axis, Cryoglobulin level in
Cryoglobulinemia in Chikungunya Infection
www.plosntds.org8 February 2009 | Volume 3 | Issue 2 | e374
present in all and combined with PVD in 24%. The concomitancy
of arthralgias and cryoglobulinemia in more than 90% of the cases
is consistent, although not demonstrative, with an involvement of
cryoglobulin in arthralgias. Frequent exacerbation of rheumatic
pain and handicap, and the occasional incidence of transitory
PVD have been recently described within the 2ndand/or 3rd
months after CHIKV-infection onset . These clinical manifes-
tations are synchronous with increasing cryoglobulin levels. The
low cryoglobulin blood levels -when compared with chronic HCV
infection- could explain the lack of vascular purpura in our cohort.
No significant association between PVD and CHIKV-MC
prevalence was observed, possibly due to the small number of
patients, although there was a positive trend between presence of
PVD and CHIKV-MC levels (p=0.06). We failed to find any
significant association between CHIKV-MC type or level and any
other sign or symptom.
Most patients self-declared improvements after short-term
general corticotherapy. In few of them we could evidence
concomitant CHIKV-MC disappearance. Corticosteroids could
interfere through two actions: immunosuppression of B lympho-
cyte activation and/or decrease of joint and tendon inflammation.
Randomized studies are needed to confirm the benefits and risks of
general corticotherapy and to specify therapeutic modalities in
chronic CHIKV-associated rheumatism. Complementary investi-
gations are also required to determine whether antiviral drugs,
such as chloroquine or interferon could be helpful in stopping
CHIKV replication in muscle satellite cells and MC production in
symptomatic chronic patients, as suggested elsewhere .
Finally, the present work also shows the evolution of CHIKV-
MC types and levels over time. Schematically, type III or type II
cryoglobulin appear first (median delay: around 40 days), followed
by type II–III (median delay: around 2 months), conversely to
what is usually described in HCV infection . Cryoglobulin
levels reached their acme within the 3rd month after disease onset
and remained stable up to the 6th, before a slow decline.
Regarding the parallel evolution of CHIKV-MC and anti-
CHIKV IgM antibodies over time and the systematic detection
of IgM in CHIKV-MC, the direct involvement of these antibodies
in cryoprecipitates can be suspected. After 6 months of evolution,
CHIKV-MC levels decrease for most patients, in parallel with
obvious clinical improvement. No cryoglobulin was detected in
most of patients after they became symptom-free, whatever the
delay of recovery. However the evolution was not linear. About
one quarter of the patients underwent clinical relapses with distal
arthralgias and a concomitant increase or reappearance of
cryoglobulin. The disappearance of CHIKV-MC could be
predictive for a clinical cure in chronic CHIKV disease, as
described in HCV infection after interferon and ribavirin or
rituximab treatment . Further studies with long time follow-up
are required to determine if type or level of early CHIKV-MC has
a real prognostic value. In Brighton’s experience, 87.9% of 107
CHIKV-infected patients self-declared cured 3 years after disease
Figure 7. Individual follow up of a 66-year-old female patient with CHIKV infection. X-axis, month of follow-up; Y-axis, Cryoglobulin level
Cryoglobulinemia in Chikungunya Infection
www.plosntds.org9 February 2009 | Volume 3 | Issue 2 | e374
onset, while 12.1% mentioned persistent symptoms including
occasional discomfort, persistent joint stiffness, or stiffness and pain
and effusion (3.7%, 2.8%, and 6% respectively) .
CHIKV infection is currently spreading in Africa, Indian
Ocean, India and Southeastern Asia and threats many others
areas where Aedes spp. is present. It is responsible for long-persisting
symptoms, which severely impair quality of life of CHIKV-
infected patients although natives or travelers. Thus, the
identification of very high prevalence of CHIKV-MC is of
importance. First, its presence can induce false negativity in
serology performed at room temperature; leading to the
recommendation of using pre-warmed sera for serology both for
individual diagnosis, and seroprevalence estimation in endemic
areas. Second, CHIKV-MC may be involved in the pathogenesis
of the chronic stage, mainly CHIKV-associated chronic rheuma-
tism and PVD. Third, its prolonged disappearance could be a
marker of the definitive clinical cure.
This work was supported by the French Service de Sante ´ des Arme ´es. We
are grateful to Mrs Mathilde Guionnet and Dr Pierre Hance for their
technical support. Preliminary results of this study were presented at the
16thEuropean Congress of Clinical Microbiology and Infectious Diseases
on April 1–4, Nice, France, and at the XIIe `mesActualite ´s du Pharo et de l’Ho ˆpital
Laveran, Marseille, France, September 7–9, 2006.
Conceived and designed the experiments: MO FS. Performed the
experiments: MO EBN HT FS. Analyzed the data: CM CR. Contributed
reagents/materials/analysis tools: MO MG HT JLM PK MM JJM PJ.
Wrote the paper: MO HT PP FS.
1. Simon F, Parola P, Grandadam M, Fourcade S, Oliver M, et al. (2007)
Chikungunya infection: an emerging rheumatism among travelers returned from
Indian Ocean islands. Report of 47 cases. Medicine (Baltimore) 86: 123–137.
2. Ross RW (1956) The Newala epidemic. III. The virus: isolation, pathogenic
properties and relationship to the epidemic. J Hyg (Lond) 54: 177–191.
3. Fourie ED, Morrison JG (1979) Rheumatoid arthritic syndrome after
chikungunya fever. S Afr Med J 56: 130–132.
4. Brighton SW, Prozesky OW, de la Harpe AL (1983) Chikungunya virus
infection. A retrospective study of 107 cases. S Afr Med J 63: 313–315.
5. Mackenzie JS, Chua KB, Daniels PW, Eaton BT, Field HE, et al. (2001)
Emerging viral diseases of Southeast Asia and the Western Pacific. Emerg Infect
Dis 7: 497–504.
6. Pastorino B, Muyembe-Tamfum JJ, Bessaud M, Tock F, Tolou H, et al. (2004)
Epidemic resurgence of Chikungunya virus in democratic Republic of the
Congo: identification of a new central African strain. J Med Virol 74: 277–282.
7. Charrel RN, de Lamballerie X, Raoult D (2007) Chikungunya outbreaks–the
globalization of vectorborne diseases. N Engl J Med 356: 769–771.
8. Josseran L, Paquet C, Zehgnoun A, Caillere N, Le Tertre A, et al. (2006)
Chikungunya disease outbreak, Reunion Island. Emerg Infect Dis 12:
9. Mavalankar D, Shastri P, Raman P (2007) Chikungunya epidemic in India: a
major public-health disaster. Lancet Infect Dis 7: 306–307.
10. Arankalle VA, Shrivastava S, Cherian S, Gunjikar RS, Walimbe AM, et al.
(2007) Genetic divergence of Chikungunya viruses in India (1963–2006) with
special reference to the 2005–2006 explosive epidemic. J Gen Virol 88:
11. Outbreak news. Chikungunya and dengue, south-west Indian Ocean. Wkly
Epidemiol Rec 81: 106–108.
12. Hochedez P, Hausfater P, Jaureguiberry S, Gay F, Datry A, et al. (2007) Cases of
chikungunya fever imported from the islands of the South West Indian Ocean to
Paris, France. Euro Surveill 12.
13. Chikungunya fever diagnosed among international travelers–United States,
2005–2006. MMWR Morb Mortal Wkly Rep 55: 1040–1042.
14. Update: chikungunya fever diagnosed among international travelers–United
States, 2006. MMWR Morb Mortal Wkly Rep 56: 276–277.
15. Parola P, de Lamballerie X, Jourdan J, Rovery C, Vaillant V, et al. (2006) Novel
chikungunya virus variant in travelers returning from Indian Ocean islands.
Emerg Infect Dis 12: 1493–1499.
16. Watson R (2007) Chikungunya fever is transmitted locally in Europe for first
time. Bmj 335: 532–533.
17. Borgherini G, Poubeau P, Staikowsky F, Lory M, Le Moullec N, et al. (2007)
Outbreak of chikungunya on Reunion Island: early clinical and laboratory
features in 157 adult patients. Clin Infect Dis 44: 1401–1407.
18. Simon F, Paule P, Oliver M (2007) Chikungunya-induced myopericarditis
toward an increase of dilated cardiomyopathy in epidemic countries? Am J Trop
19. Tedeschi A, Barate C, Minola E, Morra E (2007) Cryoglobulinemia. Blood Rev
20. Dispenzieri A, Gorevic PD (1999) Cryoglobulinemia. Hematol Oncol Clin
North Am 13: 1315–1349.
21. Pastorino B, Bessaud M, Grandadam M, Murri S, Tolou HJ, et al. (2005)
Development of a TaqMan RT-PCR assay without RNA extraction step for the
detection and quantification of African Chikungunya viruses. J Virol Methods
22. Oliver M, Coton T, Ragot C, Chianea D, Moalic JL, et al. (2005)
[Cryoglobulins: detection, type determination and quantitation. A study of
healthy subjects and patients with chronic hepatitis C]. Ann Biol Clin (Paris) 63:
23. Musset L, Diemert MC, Taibi F, Thi Huong Du L, Cacoub P, et al. (1992)
Characterization of cryoglobulins by immunoblotting. Clin Chem 38: 798–802.
24. Brouet JC, Clauvel JP, Danon F, Klein M, Seligmann M (1974) Biologic and
clinical significance of cryoglobulins. A report of 86 cases. Am J Med 57:
25. Laurent P, Le Roux K, Grivard P, Bertil G, Naze F, et al. (2007) Development
of a sensitive real-time reverse transcriptase PCR assay with an internal control
to detect and quantify chikungunya virus. Clin Chem 53: 1408–1414.
26. Agnello V, Chung R, Kaplan L (1992) A role for hepatitis C virus infection in
type II cryoglobulinemia. N engl J Med 327: 1490–1495.
27. Cacoub P, Lunel F, Musset L, Perrin M, Frangeul L, et al. (1994) Mixed
cryoglobulinemia and hepatitis C virus. Am J Med 96: 124–131.
28. Ozden S, Huerre M, Riviere JP, Coffey LL, Afonso PV, et al. (2007) Human
muscle satellite cells as targets of chikungunya virus infection. PLoS ONE 2:
29. Jaffar-Bandjee M-C, Hoareau J-J, Krejbich Trotot P, Das T, Jaffar-Bandjee Z,
et al. (3–4 december 2007) Identification des cellules cibles et des me ´canismes
physiopathologiques du virus Chikungunya. chikungunya et autres arboviroses
e ´mergentes en milieu tropical. Saint Pierre La Re ´union Island.
30. Casato M, Lilli D, Donato G, Granata M, Conti V, et al. (2003) Occult hepatitis
C virus infection in type II mixed cryoglobulinaemia. J Viral Hepat 10: 455–459.
31. Sourisseau M, Schilte C, Casartelli N, Trouillet C, Guivel-Benhassine F, et al.
(2007) Characterization of Reemerging Chikungunya Virus. PLoS Pathog 3:
32. Cicardi M, Cesana B, Del Ninno E, Pappalardo E, Silini E, et al. (2000)
Prevalence and risk factors for the presence of serum cryoglobulins in patients
with chronic hepatitis C. J Viral Hepat 7: 138–143.
Cryoglobulinemia in Chikungunya Infection
www.plosntds.org10 February 2009 | Volume 3 | Issue 2 | e374