ARTHRITIS & RHEUMATISM
Vol. 44, No. 8, August 2001, pp 1836–1840
A New Type of Acquired C1 Inhibitor Deficiency Associated
With Systemic Lupus Erythematosus
Patrice Cacoub,1 Véronique Frémeaux-Bacchi,2 Isabelle De Lacroix,3 Françoise Guillien,4
Marcel-Francis Kahn,5 Michel D. Kazatchkine,2 Pierre Godeau,1 and Jean Charles Piette1
Acquired C1 inhibitor (C1-INH) deficiency with
consequent angioedema is a rare condition that may
indicate an underlying lymphoproliferative disorder.
The defect is caused by increased catabolism, which is
often associated with the presence of serum autoanti-
bodies to C1-INH. The present report describes 3 pa-
tients with systemic lupus erythematosus who developed
typical symptoms of acquired angioedema, character-
ized by recurrent swelling of subcutaneous and mucous
tissues. The 3 patients demonstrated a major classical
pathway–mediated complement consumption, with very
low levels of C3 antigen and decreased levels of C1-INH
antigen. Neither antibodies to C1-INH nor associated
lymphoproliferative disease was found. No patient had
clinical and biologic signs of lupus activity at the time
the angioedema occurred. All patients were treated with
steroids and exhibited a good response, without relapse
of angioedema and with normalization of plasma levels
of C1-INH. In lupus patients who present with an
angioedema syndrome, acquired or hereditary angio-
edema must be sought by examining parameters of the
classical pathway and levels of C1-INH. Our observa-
tions suggest the existence of a new form of acquired
C1-INH deficiency associated with a major classical
pathway–mediated complement consumption and sys-
temic autoimmunity.
1
Patrice Cacoub, MD, Pierre Godeau, MD, Jean Charles
Piette, MD: Hôpital La Pitié-Salpêtrière, Paris, France; 2Véronique
Frémeaux-Bacchi, MD, PhD, Michel D. Kazatchkine, MD, PhD:
Hôpital Européen Georges Pompidou, Paris, France; 3Isabelle De
Lacroix, MD: Hôpital Intercommunal de Créteil, Créteil, France;
4
Françoise Guillien, MD: Hôpital Sud Francilien, Evry, France;
5
Marcel-Francis Kahn, MD: Hôpital Bichat, Paris, France.
Address correspondence and reprint requests to Professor
Patrice Cacoub, MD, Department of Internal Medicine, Hôpital La
Pitié-Salpêtrière, 83 Boulevard de l’Hôpital, 75651 Cedex 13 Paris,
France.
Submitted for publication November 3, 2000; accepted in
revised form March 27, 2001.
1836
© 2001, American College of Rheumatology
Published by Wiley-Liss, Inc.
Clinical manifestations associated with C1 inhib-
itor (C1-INH) deficiency include acute episodes of
swelling of the extremities, face, trunk, airways, and
abdominal viscera. The associated laryngeal edema may
be life-threatening. C1-INH deficiency is most often an
inherited disease that is transmitted in an autosomal-
dominant manner. Two forms of hereditary angioedema
(HAE) have been defined by immunochemical and
functional studies of the C1-INH molecule (1). Type 1
HAE, the predominant form, has low antigenic and
functional levels of C1-INH. Type 2 HAE has normal or
elevated antigenic levels and low functional levels of
C1-INH. Both types of HAE are further characterized
by the frequent onset of symptoms during childhood or
early adolescence and usually, but not always, a family
history of the disorder. Recently, a third type of HAE
that is associated with normal C1-INH activity and
occurs exclusively in women has been reported (2).
An acquired form of C1-INH with symptoms
similar to those of HAE was recognized in 1972 and is
now identified as acquired angioedema (AAE). The
onset of symptoms of AAE takes place after the fourth
decade of life in the absence of a symptomatic family
history. Two forms of AAE have been defined (3). The
first form is usually associated with B cell lymphoprolif-
erative disorders, such as lymphoma and chronic lym-
phocytic leukemia, and the absence of detectable serum
autoantibodies to C1-INH. The second form is associ-
ated with the presence of autoantibodies to C1-INH (4).
Herein we present the cases of 3 patients with
systemic lupus erythematosus (SLE) who developed a
transient angioedema associated with biologic manifes-
tations of C1-INH deficiency. In the absence of a
lymphoproliferative disorder and of autoantibodies to
C1-INH, our observations suggest the existence of a new
type of acquired C1-INH deficiency in the context of
systemic autoimmunity.
C1 INHIBITOR DEFICIENCY AND SLE
PATIENTS AND METHODS
The 3 patients included in the present study were seen
during the period 1991–1999. The patients met the criteria of
the American College of Rheumatology for the diagnosis of
SLE (5).
Complement and C1-INH assays. Plasma concentra-
tions of C1-INH, C4, and C3 were measured by nephelometry
(Beckman, Gagny, France). Normal values established with
plasma from 100 healthy donors were in the range of 12 ⫾ 3
mg/dl, 22 ⫾ 12 mg/dl, and 85 ⫾ 25 mg/dl (mean ⫾ 2SD) for
C1-INH, C4, and C3, respectively. CH50 and C2 hemolytic
activity were determined according to standard procedures (6).
Results were expressed as the percentage of the CH50 and C2
activities of a reference plasma pool obtained from 100 healthy
donors.
Antibodies to C1-INH were detected by enzyme-linked
immunosorbent assay (ELISA) as described by Alsenz et al
(4). Briefly, plastic plates (Maxisorp; Nunc, Roskilde, Den-
mark) were coated with purified functional C1-INH (Baxter,
Maurepas, France) overnight at 4°C. Free sites were blocked
with phosphate buffered saline (PBS) containing 1.0% gelatin.
After washing, plasma samples diluted in PBS/gelatin were
added to the plates. After washing, immunoglobulins bound to
C1-INH were detected by class-specific peroxidase-labeled
anti-human immunoglobulin antibodies (Biosys, Compiègne,
France), followed by orthophenylenediamine. Plasma from 1
patient with type 2 acquired C1-INH deficiency, a 1:1,000
dilution of which gave half-maximum binding, was used as an
internal standard for each class-specific immunoglobulin (IgG,
IgA, or IgM). The level in this plasma sample was arbitrarily
set at 1,000 arbitrary units (AU). Titers higher than the
mean ⫹ 2SD of those obtained in a group of 50 healthy
controls (⬎20 AU/ml) were considered positive.
Western blotting of C1-INH. Plasma was diluted in
sodium dodecyl sulfate (SDS) buffer and, in order to load 12
ng of C1-INH protein, on a 7.5% polyacrylamide gel. After
electrophoresis, proteins were transferred onto nitrocellulose
membranes. After blocking with PBS containing 5% nonfat
milk, the membrane was incubated with goat anti-human
C1-INH peroxidase conjugate (The Binding Site, Birmingham,
UK) and developed with chloronaphthol. The relative molec-
ular weights of the bands were estimated from high molecular
weight protein markers (Bio-Rad, Richmond, CA). Native
C1-INH appeared as a single band of 106 kd and cleaved
C1-INH as a single band of 97 kd.
Determination of IgG anti–collagen-like region/C1q
(anti-CLR/C1q) antibodies. Antibodies to CLR/C1q were
quantitated by ELISA using purified human CLR as antigen
(7). Briefly, each ELISA plate contained a reference positive
plasma sample (with an arbitrary value of 1,000 AU/ml) from
a patient with hypocomplementemic urticarial vasculitis. Titers
higher than the mean ⫹ 3SD of those obtained in a group of
128 healthy controls (⬎425 AU/ml) were considered positive.
CASE REPORTS
Patient 1. In 1973, this patient, a 25-year-old
woman, had arthralgia, lupus glomerulonephritis on
renal biopsy (membranous glomerulonephritis; World
1837
Health Organization [WHO] grade V), positive findings
on a test for antinuclear antibody (ANA), and high-titer
antibody to double-stranded DNA (anti-dsDNA). She
was diagnosed as having SLE and started on hydroxy-
chloroquine (400 mg/day). Low-dose prednisone (10
mg/day) was added to her therapy in 1982. Methotrexate
(15 mg/week) was added in 1986 to treat arthritis. In
January 1993, she developed a laryngeal angioedema.
During the following months, she experienced
additional episodes of subcutaneous and laryngeal an-
gioedema in the absence of clinical evidence of SLE
activity. Immunologic evaluation demonstrated a major
classical pathway–mediated complement consumption
(CH50 33%, C3 31 mg/dl, C4 ⬍3 mg/dl) and evidence of
alternative pathway activation (factor B 16 mg/dl). The
plasma level of C1-INH was reduced below the normal
range (7 mg/dl). There was no detectable anti–C1-INH
antibody of either the IgG, IgM, or IgA isotype. Evalu-
ation for a lymphoproliferative disorder yielded negative
results. Her plasma titer of anti-CLR/C1q antibodies
was high at 1,031 AU/ml.
Treatment with danazol was started in November
1993, and after 15 days, her C1-INH level had returned
to normal and the angioedema had resolved. In Decem-
ber 1993, danazol was stopped because of hepatic toxi-
city. In April 1994, the patient developed a laryngeal
angioedema after treatment with norfloxacin for a uri-
nary tract infection. In July 1994, the prednisone dosage
was increased to 40 mg/day for lupus pleuritis. At that
time, the plasma level of C1-INH was low, and a major
classical pathway– and alternative pathway–mediated
complement consumption was evidenced. During the
years that have followed, the patient has had neither
SLE flares nor angioedema. Plasma levels of classical
complement components and C1-INH have remained
within the normal range.
Patient 2. In 1993, this patient, a 16-year-old girl,
had a fever, a butterfly rash, arthralgia, positive findings
on a test for ANA, high-titer antibody to dsDNA,
antiphospholipid antibody with venous thrombosis, and
thrombopenia, which led to a diagnosis of SLE. Pred-
nisone (0.5 mg/kg/day) was started in July 1993. Two
days later, she developed labial and tracheal angio-
edema, and the prednisone was stopped. Immunologic
evaluation demonstrated a major classical pathway–
mediated complement consumption (CH50 30%, C3
⬍20 mg/dl, C4 ⬍3 mg/dl). There was also evidence of
alternative pathway activation (factor B 16 mg/dl). Her
plasma level of C1-INH was reduced below the normal
range (6 mg/dl), in the absence of detectable anti–C1-
INH antibodies. Her plasma titer of anti-CLR/C1q
1838
Table 1. Characteristics of 3 female patients with SLE and AAE*
Patient 1
Age, years† 45
ACR criteria for SLE diagnosis Glomerulonephritis
Arthralgia
Pleuritis
ANA
Anti-dsDNA
antibodies
Angioedema Yes
SLE activity at the time of AAE No
Delay between the onset of 20 years
SLE and the onset of AAE
Effect of steroids on AAE Positive
* SLE ⫽ systemic lupus erythematosus; AAE ⫽ acquired angioedema; ACR ⫽ American College of
Rheumatology; ANA ⫽ antinuclear antibodies; anti-dsDNA ⫽ anti–double-stranded DNA.
† At diagnosis of hereditary angioedema.
antibody was high at 1,344 AU/ml. Evaluation for a
lymphoproliferative disorder yielded negative results,
and there was no family history of HAE.
Prednisone treatment was gradually resumed in
October 1993, without the occurrence of cutaneous
manifestations. By March 1994, the clinical angioedema
had resolved. Plasma levels of classical and alternative
complement pathway components had increased, al-
though they remained reduced compared with normal
levels. Her plasma concentration of C1-INH was nor-
mal.
Patient 3. This patient, a 45-year-old woman, had
fever, inflammatory joint pain, photosensitivity, and
recurrent episodes of nonpruritic swelling of her legs
and arms since 1980. The patient’s sister had died of
lupus nephritis several years previously. In 1988, the
patient developed laryngeal angioedema after a thyroid-
ectomy which required a tracheotomy. Thereafter, she
had recurrent episodes of angioedema of the face and
tongue. In May 1993, an immunologic evaluation dem-
onstrated a major classical pathway–mediated comple-
ment consumption (CH50 20%, C3 21 mg/dl, C4 ⬍3
mg/dl). There was also evidence of alternative pathway
activation (factor B ⬍13 mg/dl). Her plasma level of
C1-INH was reduced below the normal range (8 mg/dl),
in the absence of detectable anti–C1-INH antibodies.
The anti-CLR/C1q antibody titer was 893 AU/ml. Eval-
uation for a lymphoproliferative disorder yielded nega-
tive results.
Treatment with prednisone (0.5 mg/kg/day) was
initiated, and 1 month later, the angioedema had re-
solved. Her levels of classical pathway complement
components were still reduced, but the C1-INH level
had increased to normal. In September 1994, she was
CACOUB ET AL
Patient 2 Patient 3
16 45
Arthralgia Glomerulonephritis
Cutaneous lesions Arthralgia
– –
ANA ANA
Anti-dsDNA Anti-dsDNA
antibodies antibodies
Yes Yes
No No
Concomitant Concomitant
Positive Positive
hospitalized for nephrotic syndrome. She was positive
for ANA and anti-dsDNA antibodies. A renal biopsy
showed a proliferative diffuse SLE glomerulonephritis
(WHO grade IV). She was treated with prednisone (1
mg/kg/day) and cyclophosphamide, and the manifesta-
tions of angioedema resolved and the C1-INH level
remained normal. One month later, she stopped treat-
ment and died unexpectedly at home.
At the time the angioedema occurred, none of
the 3 patients had received any treatment known to
cause angioedema, such as oral contraceptives, nonste-
roidal antiinflammatory drugs, or angiotensin-
converting enzyme inhibitors.
RESULTS
The main clinical features of the patients are
summarized in Table 1. All patients were female, and all
presented with typical features of SLE. Their mean ⫾
SD age was 36.4 ⫾ 16.4 years (age range 16–45). The 3
patients had positive results on tests for ANA and
high-titer anti-dsDNA antibodies; 2 patients had lupus
glomerulonephritis. The 3 patients exhibited character-
istic clinical manifestations of angioedema: nonpruritic,
nonerythematous, nonpitting, and nonpainful subcuta-
neous or mucous edema. None of the patients had
clinical or biologic signs of lupus activity at the time of
occurrence of the angioedematous crisis. The diagnoses
of SLE and angioedema were concomitantly made for
patients 2 and 3. The diagnosis of angioedema was made
20 years after the onset of SLE in the case of patient 1.
All 3 patients responded to prednisone, without a clini-
cal relapse of angioedema, within 15–36 months of
followup.
C1 INHIBITOR DEFICIENCY AND SLE
Table 2. Immunologic investigations before and after steroid treatment for acquired angioedema*
Patient 1
Before After
treatment treatment treatment treatment treatment treatment
CH50, normal 70–130% 33 31
C3 antigen, normal 60–110 mg/dl 31 40
C4 antigen, normal 12–32 mg/dl ⬍3 7 ⬍3
Factor B, normal 16–40 mg/dl 16 23
C1-INH, normal 9–15 mg/dl 7 13
C1-INH autoantibody ⫺ ⫺
IgG anti-CLR/C1q antibodies ⫹ ⫹
* C1-INH ⫽ C1 inhibitor; anti-CLR/C1q ⫽ anti–collagen-like region/C1q.
The results of immunologic investigations are
summarized in Table 2. All patients demonstrated a
major classical pathway–mediated complement con-
sumption, as evidenced by very low levels of C3 and C4,
and undetectable C2 activity. All 3 patients had a
C1-INH level below the normal range. Immunoglobulins
of different classes (IgG, IgA, or IgM) that are able to
bind normal C1-INH coated onto microtiter plates were
not detectable. SDS–polyacrylamide gel electrophoresis
analysis of C1-INH protein was performed on all 3
patients, 5 other SLE patients without evidence of
C1-INH deficiency, and 5 healthy control subjects. All
tested samples showed a C1-INH protein band at ⬃105
kd in the absence of a lower molecular weight band.
High titers of anti-CLR/C1q antibodies of the IgG
isotype were present in the plasma samples from all 3
study patients. Evaluations for a lymphoproliferative
disorder yielded negative results in all 3 patients. C1-
INH levels returned to normal with steroid treatment.
DISCUSSION
In the present study, we report 3 cases of SLE
associated with AAE. The diagnosis of C1-INH defi-
ciency was based on typical clinical manifestations and
low plasma levels of C1-INH antigen. The diagnosis of
AAE was based on the remission of angioedema with
steroid treatment and normalization of C1-INH levels in
all patients. Two forms of AAE have been described so
far. One form is associated with the presence of auto-
antibodies directed against C1-INH (8,9). This particu-
lar form of AAE has not previously been described in
patients with SLE. All sera tested from the 3 patients
described herein were negative for C1-INH autoanti-
bodies. In addition, AAE associated with C1-INH anti-
bodies is characterized by the presence of cleaved forms
of C1-INH with a circulating fragment of 96 kd. Western
blot analysis of C1-INH in 2 patients in whom the test
1839
Patient 2 Patient 3
Before After Before After
30 26 20 30
⬍20 28 21 26
5 ⬍3 ⬍3
16 20 ⬍13 15
6 10 8 11
⫺ ⫺ ⫺ ⫺
⫹ ⫹ ⫹ ⫹
could be done, demonstrated the exclusive presence of
C1-INH of normal molecular weight (105 kd).
The other form of AAE described so far is
associated with lymphoproliferative disorders with an
acquired C1-INH deficiency due to increased consump-
tion or destruction of C1-INH. It has previously been
suggested that the antibody that binds to circulating
C1-INH also reacts with an epitope of the monoclonal
immunoglobulin expressed at the surface of the B cell or
in the cytoplasm of marrow pre–B cells. The search for
a lymphoproliferative disorder gave negative findings in
all 3 patients. The 2 patients who are still alive several
years after the first angioedematous manifestations have
not developed a lymphoproliferative disorder.
Immunologic investigations of C1-INH defi-
ciency usually show decreased levels of C4 and C2, and
normal levels of C3. Quantitative and functional mea-
surements of C1-INH confirm the diagnosis. The 3
patients described herein exhibited a major classical
pathway–mediated complement consumption with very
low plasma levels of C3. Major classical pathway–
mediated complement consumption is not rare in pa-
tients with SLE; it is only poorly correlated with disease
activity. Decreased C1-INH levels may be explained by
an increased consumption of C1-INH during major
classical pathway–mediated complement activation.
It remains to be determined how the severe
classical pathway–mediated complement consumption in
our SLE patients set off their angioedema. At this time,
the presence of an autoantibody of unknown specificity
that was responsible for triggering the C1-INH con-
sumption and that is different from the autoantibodies
found in AAE cannot be ruled out. The severe comple-
ment consumption that was observed in all 3 of our
patients was associated with high titers of autoantibodies
against C1q. Thirty percent to 50% of all SLE patient
sera contain IgG antibodies reactive with human C1q
1840
(8). Most, if not all, of the IgG that is reactive with C1q
in SLE sera consists of antibodies that recognize the
collagen-like region of the molecule. C1q autoantibodies
are associated with lupus nephritis, classical pathway
activation, and high titers of antibodies to dsDNA; the
C1q antibody titers, however, do not correlate with
general disease activity (10).
Although it is a rare disorder, acquired C1-INH
deficiency has previously been described during systemic
lupus with functional and, in a few cases, with quantita-
tive C1-INH deficiency (11–13). The association of a
clinical history of recurrent AAE has been reported in
only a few patients, whereas cases of systemic or discoid
lupus erythematosus associated with HAE have been
described (14). The 3 patients described in this report
presented with typical clinical history of acquired recur-
rent, nonpitting, and nonpruritic angioedema involving
the skin and associated with classical pathway–mediated
hypocomplementemia, IgG anti-CLR/C1q antibody, and
a severe form of SLE. AAE and SLE is a rare associa-
tion, is not easy to diagnose, and perhaps, is underevalu-
ated in patients in whom steroid treatment is successful.
Taken together, our observations thus suggest a new
clinical form of acquired C1-INH deficiency associated
with systemic autoimmunity and major classical
pathway–mediated complement consumption.
REFERENCES
1. Agostoni A, Cicardi M. Hereditary and acquired C1-inhibitor
deficiency: biological and clinical characteristics in 235 patients.
Medicine (Baltimore) 1992;71:206–15.
CACOUB ET AL
2. Bork K, Barnstedt S, Koch P, Traupe H. Hereditary angioedema
with normal C1-inhibitor activity in women. Lancet 2000;356:
213–7.
3. Gelfand J, Boss G, Conley C, Reinhart R, Franck M. Acquired C1
esterase inhibitor deficiency and angioedema: a review. Medicine
(Baltimore) 1979;58:321–8.
4. Alsenz J, Bork K, Loos M. Autoantibody-mediated acquired
deficiency of C1 inhibitor. N Engl J Med 1987;316:1360–6.
5. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield
NF, et al. The 1982 revised criteria for the classification of systemic
lupus erythematosus. Arthritis Rheum 1982;25:1271–7.
6. Kazatchkine MD, Hauptmann G, Nydegger V. Techniques du
complément. Paris: Editions INSERM; 1985.
7. Frémeaux-Bacchi V, Weiss L, Demouchy C, Blouin J, Kazatchkine
MD. Autoantibodies to the collagen-like region of C1q are
strongly associated with classical pathway-mediated hypocomple-
mentemia in systemic lupus erythematosus. Lupus 1996;5:216–20.
8. Whaley K, Sim R, He S. Autoimmune C1-inhibitor deficiency. Clin
Exp Immunol 1996;106:423–6.
9. Markovic S, Inwards D, Frigas E, Phyliky R. Acquired C1 esterase
inhibitor deficiency. Ann Intern Med 2000;132:144–50.
10. Siegert C, Daha M, Westedt M, van der Voort E, Breedveld F. IgG
autoantibodies against C1q are correlated with dsDNA antibodies,
hypocomplementemia and nephritis in systemic lupus erythema-
tosus. J Rheumatol 1991;18:230–4.
11. Jazwinska E, Gatenby P, Dunckley H, Serjeantson S. C1 inhibitor
functional deficiency in systemic lupus erythematosus (SLE). Clin
Exp Immunol 1993;92:268–73.
12. Ochonisky S, Intrator L, Wechsler J, Revuz J, Bagot M. Acquired
C1 inhibitor deficiency revealing systemic lupus erythematosus.
Dermatology 1993;186:261–3.
13. Fawaz-Estrup F. Human parvovirus infection: rheumatic manifes-
tations, angioedema, C1 esterase inhibitor deficiency, ANA posi-
tivity, and possible onset of systemic lupus erythematosus. J Rheu-
matol 1996;23:1180–5.
14. Donaldson V, Hess E, McAdams A. Lupus erythematosus-like
disease in three unrelated women with hereditary angioneurotic
edema. Ann Intern Med 1977;86:312–3.