Accepted Manuscript

Intravenous immunoglobulins improve survival in monoclonal gammopathy-

associated systemic capillary-leak syndrome

Marc Pineton de Chambrun, MD, Marie Gousseff, MD, Wladimir Mauhin, MD, Jean-
Christophe Lega, MD, PhD, Marc Lambert, MD, PhD, Sophie Rivière, MD, Antoine
Dossier, MD, Marc Ruivard, MD, PhD, François Lhote, MD, Gilles Blaison, MD,
Laurent Alric, MD, PhD, Christian Agard, MD, PhD, David Saadoun, MD, PhD, Julie
Graveleau, MD, Martin Soubrier, MD, PhD, Marie-Josée Lucchini-Lecomte, MD,
Christine Christides, MD, Annick Bosseray, MD, Hervé Levesque, MD, PhD, Jean-
François Viallard, MD, PhD, Nathalie Tieulie, MD, Pierre-Yves Lovey, MD, Sylvie
Le Moal, MD, Béatrice Bibes, MD, Giuseppe Malizia, MD, Pierre Abgueguen, MD,
François Lifermann, MD, Jacques Ninet, MD, PhD, Pierre-Yves Hatron, MD, Zahir
Amoura, MD, MSc
PII: S0002-9343(17)30602-2
DOI: 10.1016/j.amjmed.2017.05.023
Reference: AJM 14127

To appear in: The American Journal of Medicine

Received Date: 7 December 2016

Revised Date: 28 March 2017
Accepted Date: 9 May 2017

Please cite this article as: de Chambrun MP, Gousseff M, Mauhin W, Lega JC, Lambert M, Rivière
S, Dossier A, Ruivard M, Lhote F, Blaison G, Alric L, Agard C, Saadoun D, Graveleau J, Soubrier M,
Lucchini-Lecomte MJ, Christides C, Bosseray A, Levesque H, Viallard JF, Tieulie N, Lovey PY, Le
Moal S, Bibes B, Malizia G, Abgueguen P, Lifermann F, Ninet J, Hatron PY, Amoura Z, members of the
EurêClark Study Group, Intravenous immunoglobulins improve survival in monoclonal gammopathy-
associated systemic capillary-leak syndrome, The American Journal of Medicine (2017), doi: 10.1016/
j.amjmed.2017.05.023.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

1 Title

2 Intravenous immunoglobulins improve survival in monoclonal gammopathy-associated systemic

3 capillary-leak syndrome

PT
5 Authors

6 EurêClark study group

RI
7 Marc PINETON de CHAMBRUN1,2, MD, Marie GOUSSEFF3, MD, Wladimir MAUHIN1, MD, Jean-

Christophe LEGA4, MD, PhD, Marc LAMBERT5, MD, PhD, Sophie RIVIÈRE6, MD, Antoine DOSSIER7, MD,

SC
8

9 Marc RUIVARD8, MD, PhD, François LHOTE9, MD, Gilles BLAISON10, MD, Laurent ALRIC11, MD, PhD,

10 Christian AGARD12, MD, PhD, David SAADOUN13, MD, PhD, Julie GRAVELEAU14, MD, Martin

U
11 SOUBRIER15, MD, PhD, Marie-Josée LUCCHINI-LECOMTE16, MD, Christine CHRISTIDES17, MD, Annick

12
AN
BOSSERAY18, MD, Hervé LEVESQUE19, MD, PhD, Jean-François VIALLARD20, MD, PhD, Nathalie

13 TIEULIE21, MD, Pierre-Yves LOVEY22, MD, Sylvie LE MOAL23, MD, Béatrice BIBES24, MD, Giuseppe
M
14 MALIZIA25, MD, Pierre ABGUEGUEN26, MD, François LIFERMANN27, MD, Jacques NINET28, MD, PhD,
D

15 Pierre-Yves HATRON5, MD, Zahir AMOURA1*, MD, MSc and members of the EurêClark Study Group.
TE

16
1
17 Service de médecine interne 2, CHU La Pitié-Salpêtrière, APHP, Université Paris 6, Paris, France
2
EP

18 Service de réanimation médicale, CHU La Pitié-Salpêtrière, APHP, Université Paris 6, Paris, France
3
19 Service de médecine interne, CH Bretagne Atlantique, Vannes, France
C

4
20 Service de médecine interne et vasculaire, Centre Hospitalier Lyon Sud, Hospices Civils de Lyon,
AC

21 CHU, Pierre-Bénite, France; UMR 5558, Laboratoire de Biométrie et Biologie Évolutive, CNRS,

22 Université Claude-Bernard Lyon 1, Lyon, France.

5
23 Service de médecine interne, Hôpital Claude Huriez, CHRU Lille, France
6
24 Service de médecine interne, Hôpital Saint-Eloi, CHRU Montpellier, France
7
25 Service de médecine interne, CHU Bichat, Paris, APHP, France
8
26 Service de médecine interne, Hôpital d'Estaing, CHU de Clermont-Ferrand, France

1
 ACCEPTED MANUSCRIPT

9
1 Service de médecine interne, Hôpital Delafontaine, Saint-Denis, France
10
2 Service de médecine interne et rhumatologie, Hôpital Pasteur, Colmar, France
11
3 Service de médecine interne, Pôle digestif, CHU Purpan, Toulouse, France
12
4 Service de médecine interne, CHU Hôtel-Dieu, Nantes, France

PT
13
5 Service de médecine interne et immunologie clinique, CHU La Pitié-Salpêtrière, APHP, Paris, France
14
6 Service de médecine polyvalente, CH Saint-Nazaire, France

RI
15
7 Service de rhumatologie, Hôpital Gabriel-Montpied, CHU Clermont-Ferrand, France
16
8 Service de médecine Interne, CH Notre-Dame de la Miséricorde, Ajaccio, France

SC
17
9 Service de médecine interne, CH Avignon, France
18
10 Service de médecine interne, CHU Grenoble, Grenoble, France

U
19
11 Normandie univ, UNIROUEN, U1096, Service de médecine interne, F 76000 Rouen

12 20
AN
Service de médecine interne, Hôpital Haut-Lévêque, CHU Bordeaux Sud, Pessac, France
21
13 Service de rhumatologie, Hôpital Pasteur 2, CHU Nice, France
M
22
14 Service d’hématologie, Hôpital du Valais-Institut Central, Sion, Suisse23Service de médecine interne,
D

15 CH de Saint-Brieuc, France
24
16 Service de médecine interne, CHP Saint-Grégoire, France
TE

25
17 Divisione di Gastroenterologia, Ospedale V. Cervello, Palermo, Italy
26
18 Service des maladies infectieuses et tropicales, CHU d'Angers, France
EP

27
19 Service de médecine interne, CH Dax, France
28
20 Service de médecine interne, CHU Edouard Herriot, Lyon, France
C

21
AC

*
22 Correspondence: zahir.amoura@aphp.fr Service de Médecine Interne 2, Institut E3M, Hôpital La

23 Pitié–Salpêtrière, 47–83, bd de l’Hôpital, 75651 Paris Cedex 13, France. Tel: +33 (0)1 42 17 80 01; Fax:

24 +33 (0)1 42 17 80 22.

25

26 Abstract word count: 283

2
 ACCEPTED MANUSCRIPT

1 Text word count: 2489

2 Tables count: 7

3 Figures count: 3

4 References count: 26

PT
5

6 Disclosures: None.

RI
7

8 Funding source: None.

SC
9

10 Role in writing the manuscript: M.PdC, M.G, W.M and Z.A designed the study, collected and

U
11 analyzed data. M.PdC and Z.A wrote the article. All other members of the EurêClark study group took

12
AN
care of patients, referred them to the EurêClark coordinating center and participated to data

13 collection.
M
14
D

15 Type of article: Clinical Research study

16
TE

17 Key words: Intravenous immunoglobulins, systemic capillary-leak syndrome, Clarkson disease,

18 monoclonal gammopathy-associated systemic capillary-leak syndrome

EP

19

20 Running head: IVIg improve survival in MG-SCLS.

C

21
AC

22 Abbreviations

23 IVIg: intravenous immunoglobulins

24 MG-SCLS: monoclonal gammopathy-associated systemic capillary-leak syndrome

25 SCLS: systemic capillary-leak syndrome

26

3
 ACCEPTED MANUSCRIPT

1 Abstract

2 Background: Monoclonal gammopathy-associated systemic capillary-leak syndrome, also known as

3 Clarkson disease, is a rare condition characterized by recurrent life-threatening episodes of capillary

4 hyper-permeability in the context of a monoclonal gammopathy. This study was conducted to better

PT
5 describe the clinical characteristics, natural history, and long-term outcome of monoclonal

6 gammopathy-associated systemic capillary-leak syndrome.

RI
7 Methods: We conducted a cohort analysis of all patients included in the European Clarkson disease

8 (EurêClark) registry between January 1997 and March 2016. From diagnosis to last follow-up, studied

SC
9 outcomes (e.g., the frequency and severity of attacks, death, and evolution toward multiple

10 myeloma) and the type of preventive treatments administered were monitored every 6 months.

U
11 Results: Sixty-nine patients (M/F sex ratio 1:1; mean ± SD age at disease onset 52 ± 12 years) were

12
AN
included in the study. All patients had monoclonal gammopathy of immunoglobulin G type, with

13 kappa light chains in 47 (68%). Median (interquartile range, IQR) follow-up duration was 5.1 (2.5-9.7)
M
14 years. Twenty-four patients (35%) died after 3.3 (0.9-8) years. Fifty-seven (86%) patients received at
D

15 least one preventive treatment, including intravenous immunoglobulins (IVIg) n = 48 (73.8%),

16 theophylline n = 22 (33.8%), terbutaline n = 22 (33.8%), and thalidomide n = 5 (7.7%). In the 65

TE

17 patients with follow-up, 5- and 10-year survival rates were 78% (n = 35) and 69% (n = 17)

18 respectively. Multivariate analysis found preventive treatment with IVIg [hazard ratio (HR) 0.27 (0.10-
EP

19 0.70), p=0.007] and terbutaline [HR 0.35 (0.13-0.96), p=0.041] to be independent predictors of

20 mortality.
C

21 Conclusions: We describe the largest cohort to date of patients with well-defined monoclonal
AC

22 gammopathy-associated systemic capillary-leak syndrome. Preventive treatment with IVIg was the

23 strongest factor associated with survival, suggesting the use of IVIg as the first line in prevention

24 therapy.

25

26

4
 ACCEPTED MANUSCRIPT

3 Introduction

4 Monoclonal gammopathy-associated systemic capillary-leak syndrome, also known as Clarkson

PT
5 disease, is a rare condition characterized by recurrent life-threatening episodes of capillary hyper-

6 permeability in the context of a monoclonal gammopathy1. Since the initial description of the disease

RI
7 by Clarkson et al. in 19601, fewer than 250 cases have been reported worldwide2. During acute

8 episodes, fluid and protein leakage from the intravascular compartment into the interstitium causes

SC
9 clinical signs of acute hypovolemia and interstitial edema. The laboratory work-up is pathognomonic,

10 with marked hemoconcentration and paradoxical hypoproteinemia1,3,4. Diagnosis relies on recurring

U
11 typical flares associated with monoclonal gammopathy (reported in >85% of patients)1,4–7, after

12
AN
exclusion of differential diagnoses of secondary capillary-leak syndrome or hypoproteinemia.

13 Optimal management remains unclear, although in recent years, a growing body of evidence
M
14 suggested that intravenous immunoglobulins (IVIg) might prevent attack recurrence5,8–15. Herein, we
D

15 report the clinical characteristics, the natural history, and the outcome of 69 patients with

16 monoclonal gammopathy-associated systemic capillary-leak syndrome.

TE

17

18 Methods
EP

19 The European Clarkson disease (EurêClark) registry

20 As previously reported5,6, the EurêClark registry is an international study group, comprising 49

C

21 medical centers in eight countries (France, Italy, Israel, Switzerland, Lebanon, Canada, Spain, and
AC

22 Turkey), which gathers monoclonal gammopathy-associated systemic capillary-leak syndrome

23 observations and prospectively monitors episodes, preventive treatments, complications, and

24 patient outcomes. In 1997, this registry was approved by local review boards (AP–HP no.14) and the

25 “Commission Nationale de l’Informatique et des Libertés” (no. 1001704). All patients, or their next of

26 kin, accepted the inclusion in the EurêClark database.

5
 ACCEPTED MANUSCRIPT

2 EurêClark inclusion and non-inclusion criteria

3 Patients were included from January 1 1997 and prospectively monitored until March 31 2016.

4 Patients could be included even when the first episode preceded the starting date for inclusions.

PT
5 Criteria for the diagnosis of Clarkson disease were as follows: 1) presence of a monoclonal

6 gammopathy; 2) one or more episodes that met all of the following criteria5: clinical signs of acute

RI
7 hypovolemia (e.g., sudden fatigue, thirst, dizziness, oliguria and low blood pressure) and acute

8 interstitial edema (e.g., myalgias, paresthesia, nausea and vomiting, abdominal pain, and generalized

SC
9 or segmental edema); 3) hemoconcentration (elevated hematocrit or hemoglobin exceeding normal

10 values for age and sex or >20% of the last reference value for a given patient) with paradoxical

U
11 hypoproteinemia; 4) and exclusion of any other cause of secondary capillary-leak syndrome or

12 hypoproteinemia2,7,16.
AN
13 In case of suspicion of Clarkson disease, patients were referred to the EurêClark coordinating
M
14 center, where medical history, clinical manifestations, and laboratory findings of every patient were
D

15 evaluated and the diagnosis confirmed or invalidated. Patients without monoclonal gammopathy

16 were not considered for inclusion in the registry, and no pediatric patients were included because
TE

17 none had monoclonal gammopathy. Some patients have been previously reported in earlier

18 EurêClark studies5,6.
EP

19

20 Data collection and definition

C

21 Standardized forms were used to collect baseline and follow-up data: epidemiologic, clinical data,
AC

22 and laboratory findings; number and frequency of attacks; and preventive treatments and outcomes.

23 After patient inclusion, referring physicians were contacted by e-mail and/or telephone every 6

24 months to complete follow-up forms.

25 Severe attacks were defined as: systolic blood pressure less than 80mmHg, mean blood pressure

26 less than 65mmHg, loss of consciousness, admission to the intensive care unit, or a combination of

6
 ACCEPTED MANUSCRIPT

1 these. Chronic evolution was defined by a high number and frequency of episodes that made them

2 non-differentiable. Monoclonal gammopathy identification and quantification were performed using

3 serum protein electrophoresis and immunofixation, always before any IVIg infusion. The diagnosis of

4 myeloma was established according to the up-to-date international guidelines available at the time

PT
5 of each evaluation.

6 Preventive treatments were defined as drugs introduced to prevent recurrence or to lower the

RI
7 severity of episodes. Four preventive treatments were considered in this study: IVIg, theophylline,

8 terbutaline, and thalidomide. According to drug tolerance, preventive treatment dosages were as

SC
9 follows: theophylline 400 mg/day to 1600 mg/day; terbutaline 15 mg/day to 25 mg/day and

10 thalidomide 50 mg/day to a maximum dosage of 100 mg/day. There was no pre-specified protocol

U
11 for the preventive treatment administration. Before year 2000, the combination of theophylline and

12
AN
terbutaline was administered as the first therapy. Since year 2000, IVIg are recommended as the first

13 line therapy. The first IVIg infusion was always initiated after the resolution of the previous attack,
M
14 because of concern regarding the safety of IVIg during severe episodes. The initial recommended IVIg
D

15 treatment was the monthly intravenous administration of 2 g/kg of body weight. We recommended

16 no specific type of IVIg and the choice depended on the preference of each center. The monthly
TE

17 treatment with 2 g/kg was administered during a minimum of 1 year. In the absence of a 1-year

18 recurrence and after the approval of the coordinating center, the IVIg treatment could be tapered to
EP

19 1 g/kg monthly and then, after another year free of recurrence, to 0.5 g/kg monthly. After achieving

20 the dosage of 0.5 g/kg monthly we considered on a case-to-case basis to increase the interval
C

21 between the IVIg infusions. In case of relapse, the IVIg treatment was returned to the previous step.
AC

22

23

24 Statistical Analysis

25 Results are expressed as frequencies (%), continuous variables as mean (standard deviation) or

26 median [interquartile range (IQR) 25-75] and compared using Student t-test or Wilcoxon rank test

7
 ACCEPTED MANUSCRIPT

1 when appropriate. Categorical variables were compared with Fisher exact tests. Patients’

2 demographic, clinical, and biological characteristics were tested in univariate analysis for association

3 with end-of-follow-up mortality. Thereafter, Cox proportional hazard model using backward-stepwise

4 variable elimination was run (with the variable exit threshold set at P>0.10). Factors achieving P less

PT
5 than or equal to0.20 in univariate analysis and parameters previously reported to be strongly

6 associated with death were entered into the multivariate model. Considering the time from diagnosis

RI
7 to death or the end of follow-up as the time to event, survival rates were estimated using Kaplan–

8 Meier curves and compared using the log rank test. Statistical significance was defined as P<0.05.

SC
9 Analyses were computed with IBM SPSS Statistics 22.0 software (IBM Corp., Armonk, NY).

10

U
11 Results

12 Patient characteristics
AN
13 Characteristics of the 69 patients included in the study, between January 1997 and March 2016, are
M
14 reported in Table 1. The male/female ratio was 1:1, with mean age at disease onset and at diagnosis
D

15 of 52 ± 12 years and 53.5 ± 12 years, respectively. All patients had an IgG monoclonal gammopathy,

16 with kappa and lambda light chains in 47 (68.1%) and 24 (34.8%) patients, respectively (2 patients
TE

17 had biclonal gammopathy). The median level of monoclonal component at diagnosis was 4.4 [2-8]

18 g/L. All patients met, at diagnosis, the criteria for monoclonal gammopathy of undetermined
EP

19 significance.
C

20

21 Outcomes
AC

22 Median length of follow-up was 5.1 [2.5-9.7] years. The median number of attacks after diagnosis

23 was 2 [0-6] including 1 [0-2] severe episodes. This number was highly heterogeneous among

24 patients, with a maximum numbers of attacks after diagnosis of 35.

25 Fifty-seven (86.2%) patients received at least one preventive treatment during follow-up,

26 including: IVIg (n = 48, 73.8%), theophylline (n = 22, 34%), terbutaline (n = 22, 34%), and thalidomide

8
 ACCEPTED MANUSCRIPT

1 (n = 5, 7.7%). Twenty-three (35.4%) patients received more than one preventive treatment. The

2 distribution of the treatments administered during follow-up is reported in Figure 1. Other outcomes

3 and preventive treatment parameters are reported in Table 2.

4 Multiple myeloma occurred in 5 (7.2%) patients. Twenty-four patients (34.8%) died after a median

PT
5 duration of 3.3 [0.9-8] years: 20 (83.3%) during a severe attack and four (16.7%) from multiple

6 myeloma. Four patients died during their first episode, and were not included in the follow-up

RI
7 analysis. One, two, five and ten-year survival rates were 97%, 95%, 78%, and 69%, respectively. The

8 Kaplan-Meier curve of probability of survival is shown in Figure 2.

SC
9

10 Factor Association with End-of-Follow-up Mortality

U
11 The comparison between survivors (n = 45) and patients who died (n = 20) during follow-up is

12
AN
reported in Table 3. Non-survivors had significantly more recurrence of severe episodes after

13 diagnosis [19 (95%) vs. 19 (42.2%), p<0.0001] and myeloma [5 (25%) vs. 0(0%), p=0.002] than did
M
14 survivors. Survivors received significantly more frequent preventive treatment with IVIg [40 (88.9%)
D

15 vs. 8 (40%), p<0.0001] than did non-survivors. According to multivariate analysis using a Cox

16 proportional model to identify mortality-associated factors (Table 4), only preventive treatment with
TE

17 IVIg [HR 0.27 (0.10-0.70), p = 0.007] and terbutaline [HR 0.35 (0.13-0.96), p = 0.04] were independent

18 predictors of mortality. Five- and 10-year survival rates in patients treated with IVIg were 91% and
EP

19 77%, respectively, compared to 47% and 37% in patients not treated with IVIg (log rank test

20 p<0.0001, Figure 3).

C

21
AC

22 Subgroup Comparison

23 Patients treated with IVIg (n = 48) had significantly less recurrence of attacks and fewer severe

24 attacks [31 (64.6%) vs. 16 (94.1%), p = 0.03 and 22 (45.8%) vs. 16 (94.1%), p<0.0001, respectively]

25 compared to patients not treated with IVIg (n = 17, Table 5).

9
 ACCEPTED MANUSCRIPT

1 Conversely, patients treated with terbutaline (n = 22) had significantly more recurrence of

2 episodes and more severe episodes [20 (90.9%) vs. 27 (62.8%), p = 0.02 and 17 (77.3%) vs. 21 (48.8),

3 p = 0.035] than patients not treated with terbutaline (n = 43, Table 6).

4 Finally, patients without severe recurrence (n = 27) received IVIg more frequently [26 (96.3%) vs.

PT
5 22 (57.9%), p<0.0001] and less often terbutaline [5 (18.5%) vs. 17 (44.7%), p=0.035] than did patients

6 with severe recurrence (n = 38, Table 7). It is worth noting that, patients with severe relapse had a

RI
7 more frequent median level of monoclonal protein at diagnosis >5 g/L [17 (44.7%) vs. 7 (25.9%), p =

8 0.006].

SC
9

10 Discussion

U
11 Monoclonal gammopathy-associated systemic capillary-leak syndrome is a puzzling condition

12
AN
responsible for recurrent shocks of unknown origin. Despite increasing evidence over the recent

13 years, data regarding long-term evolution and efficacy of preventive treatments remain scarce1.
M
14 Herein, we further characterize the disease’s clinical characteristics, long-term evolution, and
D

15 mortality-associated factors in the largest cohort reported to date.

16 The demographic characteristics of our cohort differ from those previously reported. Indeed, the
TE

17 mean age at diagnosis was over 50 years, 10 years older than other cohorts2,4,5,7,9. This age difference

18 probably reflects the exclusion of pediatric cases. Moreover, a mean age of approximately 50 years
EP

19 agrees with the natural epidemiology of monoclonal gammopathy, whose index of first flare occurs

20 during the fifth decade17,18. As part of the inclusion criteria, monoclonal gammopathy was present in
C

21 every patient, whereas rates in previous studies ranged from 76% to 82%4,7,16,19. Heavy chain was
AC

22 always an IgG, as well, in all but five previously reported patients: three had an IgA (two with typical

23 attacks20–22,22 and one with a chronic form at onset23) and two had an IgM (with chronic forms24,25).

24 Predominance of kappa light chain is concordant with previous reports2,4,16,26. Five patients

25 developed myeloma during their follow-up. Considering the number of patients in the cohort (n =

26 69), the median duration of follow-up (5 years) and a theoretical risk of progression of monoclonal

10
 ACCEPTED MANUSCRIPT

1 gammopathy toward myeloma (1% each year)17,18, the prevalence of myeloma in our cohort seems to

2 be consistent with the natural evolution of monoclonal gammopathy. There was no association

3 between the initial level of the monoclonal component and the end-of-follow-up mortality. However,

4 there was an association with the frequency of severe relapse after diagnosis. Patients with a high

PT
5 level of monoclonal component at diagnosis were more susceptible to severe relapse. This finding

6 emphasizes the role of the monoclonal gammopathy and/or the plasma cell clone in the

RI
7 pathophysiology of Clarkson disease.

8 Preventive treatments in our cohort were strongly correlated with time. Indeed, before the use of

SC
9 IVIg, the combination of terbutaline and theophylline was administered as the only preventive

10 therapy. After the year 2000, IVIg was administered as a second-line therapy, in cases of theophylline

U
11 and terbutaline failure. During the last decade, patients were treated with IVIg as a first-line therapy

12
AN
and theophylline, terbutaline, or thalidomide were only considered in cases of IVIg failure. IVIg was

13 the primary preventive treatment administered in our cohort (74%). Only one-third of our patients
M
14 received rec-agonists compared to 42% to 92% of the patients previously reported4,5.
D

15 Overall 5- and 10-year survival rates of the 65 patients with follow-up were 78% and 69%,

16 respectively, consistent with the survival rate of 24 patients from the Mayo Clinic cohort4, but slightly
TE

17 higher than the survival rate reported in two reviews (5- and 10-year survival of 70% and 66%

18 respectively)16,26.
EP

19 Importantly, preventive treatment with IVIg and terbutaline were the only factors significantly

20 associated with survival in multivariate analysis. Neither the use of thalidomide nor theophylline was
C

21 associated with improved survival. IVIg’s efficacy in preventing recurrence of episodes has been
AC

22 reported in several case reports and short series during the last decade8–15. We previously reported

23 the efficacy of a preventive treatment in reducing mortality in 28 patients5 but lacked statistical

24 power to determine which treatment between IVIg and nde-agonists was responsible for this effect.

25 Herein, we widen the evidence supporting IVIg’s efficacy on severe attacks and mortality prevention.

26 Patients treated with IVIg were more likely to be free of recurrence, severe recurrence, and alive at

11
 ACCEPTED MANUSCRIPT

1 the end of follow-up. Furthermore, all but one patient who did not experience a severe relapse were

2 treated with IVIg.

3 Multivariate analysis also considered terbutaline as an independent predictor of survival,

4 although it was not associated with survival in univariate analysis. Patients with severe recurrence

PT
5 were more likely to be treated with terbutaline than patients without severe recurrence. Moreover,

6 recurrence and severe recurrence were more frequent in patients treated with terbutaline than in

RI
7 patients not treated with terbutaline. These contradictory results could suggest the existence of a

8 confounding factor in the interpretation of terbutaline efficacy. Yet, this could also suggest that

SC
9 terbutaline did not lower recurrence but only the severity of episodes, responsible for an overall

10 effect on mortality.

U
11 Altogether, these results suggest the use of IVIg as first-line prevention therapy in patients with

12
AN
monoclonal gammopathy-associated systemic capillary-leak syndrome. Results regarding the efficacy

13 of terbutaline should be considered carefully, and terbutaline could be considered in cases of IVIg
M
14 failure.
D

15 Our study has several limitations. First, because patients were recruited over two decades,

16 treatment regimens were inevitably heterogeneous. Second, we included and compared new and old
TE

17 patients whose cases had already been published. However, this allowed us to report the largest

18 cohort to date and to increase the power of our statistical analysis. Last, we did not analyze the
EP

19 effects of other preventive treatments (e.g., calcium channel blockers, corticosteroids, Ginkgo biloba

20 extract, and pentoxifylline) in the survival analysis, which might induce a potential bias. Yet, these
C

21 treatments were only administered in a limited number of patients, over short periods of time, and
AC

22 their efficacy is highly controversial.

23

24 Conclusion

25 This study reports on a large cohort of patients with well-defined monoclonal gammopathy-

26 associated systemic capillary-leak syndrome. Five- and ten-year survival rates were 78% and 69%,

12
 ACCEPTED MANUSCRIPT

1 respectively. Multivariate analysis determined preventive treatment with IVIg as the primary factor

2 associated with survival. These results suggest the use of IVIg as the first-line preventive agent in

3 monoclonal gammopathy-associated systemic capillary-leak syndrome.

PT
5 Acknowledgment: The authors thank all the physicians caring for the patients and who referred

6 them to the registry. We also thank the patients for agreeing to participate in our registry.

RI
7

SC
9

10

U
11

12
AN
13
M
14
D

15

16
TE

17 References

18 1. Clarkson B, Thompson D, Horwith M, Luckey EH. Cyclical edema and shock due to increased
EP

19 capillary permeability. The American Journal of Medicine. 1960;29(2):193-216.

20 doi:10.1016/0002-9343(60)90018-8.
C
AC

21 2. Duron L, Delestre F, Amoura Z, Arnaud L. Syndrome de fuite capillaire idiopathique et formes

22 secondaires : une revue systématique de la littérature. La Revue de Médecine Interne.

23 2015;36(6):386-394. doi:10.1016/j.revmed.2014.11.005.

13
 ACCEPTED MANUSCRIPT

1 3. Atkinson JP, Waldmann TA, Stein SF, et al. Systemic capillary leak syndrome and monoclonal

2 IgG gammopathy; studies in a sixth patient and a review of the literature. Medicine (Baltimore).

3 1977;56(3):225-239.

4 4. Kapoor P, Greipp PT, Schaefer EW, et al. Idiopathic systemic capillary leak syndrome (Clarkson’s

PT
5 disease): the Mayo clinic experience. Mayo Clin Proc. 2010;85(10):905-912.

6 doi:10.4065/mcp.2010.0159.

RI
7 5. Gousseff M, Arnaud L, Lambert M, et al. The systemic capillary leak syndrome: a case series of

SC
8 28 patients from a European registry. Ann Intern Med. 2011;154(7):464-471. doi:10.7326/0003-

9 4819-154-7-201104050-00004.

U
10

11
6.
AN
Amoura Z, Papo T, Ninet J, et al. Systemic capillary leak syndrome: report on 13 patients with

special focus on course and treatment. Am J Med. 1997;103(6):514-519.

M
12 7. Druey KM, Greipp PR. Narrative review: the systemic capillary leak syndrome. Ann Intern Med.

13 2010;153(2):90-98. doi:10.7326/0003-4819-153-2-201007200-00005.
D
TE

14 8. Lambert M, Launay D, Hachulla E, et al. High-dose intravenous immunoglobulins dramatically

15 reverse systemic capillary leak syndrome. Crit Care Med. 2008;36(7):2184-2187.

EP

16 doi:10.1097/CCM.0b013e31817d7c71.

17 9. Xie Z, Chan EC, Long LM, Nelson C, Druey KM. High-dose intravenous immunoglobulin therapy
C

18 for systemic capillary leak syndrome (Clarkson disease). Am J Med. 2015;128(1):91-95.

AC

19 doi:10.1016/j.amjmed.2014.08.015.

20 10. Abgueguen P, Chennebault JM, Pichard E. Immunoglobulins for Treatment of Systemic Capillary

21 Leak Syndrome. The American Journal of Medicine. 2010;123(6):e3-e4.

22 doi:10.1016/j.amjmed.2009.09.034.

14
 ACCEPTED MANUSCRIPT

1 11. Almagro P, Martí JM, Garcia Pascual L, Rodriguez-Carballeira M. Successful treatment of

2 systemic capillary leak syndrome with intravenous immunoglobulins. Rev Clin Esp.

3 2012;212(4):218-219. doi:10.1016/j.rce.2011.08.004.

4 12. del Olmo Revuelto MA, Abajo del Álamo C, Santos Pérez MI. [Use of intravenous

PT
5 immunoglobulins in a patient with systemic capillary leak syndrome]. Med Clin (Barc).

6 2014;142(8):377-378. doi:10.1016/j.medcli.2013.06.021.

RI
7 13. Scanvion Q, Lefèvre G, Hachulla E, Hatron P-Y, Lambert M. Subcutaneous immunoglobulin

SC
8 therapy prevents systemic capillary leak syndrome attack. Am J Med. April 2016.

9 doi:10.1016/j.amjmed.2016.02.046.

U
10

11
14.
AN
Zipponi M, Eugster R, Birrenbach T. High-dose intravenous immunoglobulins: a promising

therapeutic approach for idiopathic systemic capillary leak syndrome. BMJ Case Rep.
M
12 2011;2011. doi:10.1136/bcr.12.2010.3599.

13 15. Shin JI, Lee JS. Beneficial effect of intravenous immunoglobulins on systemic capillary leak
D

14 syndrome in patients with monoclonal gammopathy. Crit Care Med. 2009;37(2):795; author
TE

15 reply 795. doi:10.1097/CCM.0b013e3181959c3d.

EP

16 16. Gousseff M, Amoura Z. [Idiopathic capillary leak syndrome]. Rev Med Interne. 2009;30(9):754-

17 768. doi:10.1016/j.revmed.2009.01.005.
C

18 17. Kyle RA, Therneau TM, Rajkumar SV, et al. A long-term study of prognosis in monoclonal
AC

19 gammopathy of undetermined significance. N Engl J Med. 2002;346(8):564-569.

20 doi:10.1056/NEJMoa01133202.

21 18. Therneau TM, Kyle RA, Melton LJ, et al. Incidence of monoclonal gammopathy of undetermined

22 significance and estimation of duration before first clinical recognition. Mayo Clin Proc.

23 2012;87(11):1071-1079. doi:10.1016/j.mayocp.2012.06.014.

15
 ACCEPTED MANUSCRIPT

1 19. Kawabe S, Saeki T, Yamazaki H, Nagai M, Aoyagi R, Miyamura S. Systemic Capillary Leak

2 Syndrome. Internal Medicine. 2002;41(3):211-215. doi:10.2169/internalmedicine.41.211.

3 20. Droder RM, Kyle RA, Greipp PR. Control of systemic capillary leak syndrome with aminophylline

4 and terbutaline. Am J Med. 1992;92(5):523-526.

PT
5 21. Tahirkheli NK, Greipp PR. Treatment of the systemic capillary leak syndrome with terbutaline

RI
6 and theophylline. A case series. Ann Intern Med. 1999;130(11):905-909.

SC
7 22. Xie Z, Ghosh CC, Patel R, et al. Vascular endothelial hyperpermeability induces the clinical

8 symptoms of Clarkson disease (the systemic capillary leak syndrome). Blood.

U
9 2012;119(18):4321-4332. doi:10.1182/blood-2011-08-375816.

10 23. AN
Lesterhuis WJ, Rennings AJ, Leenders WP, et al. Vascular endothelial growth factor in systemic

11 capillary leak syndrome. Am J Med. 2009;122(6):e5-e7. doi:10.1016/j.amjmed.2009.01.020.

M
12 24. Xie Z, Chan E, Yin Y, et al. Inflammatory Markers of the Systemic Capillary Leak Syndrome
D

13 (Clarkson Disease). J Clin Cell Immunol. 2014;5:1000213. doi:10.4172/2155-9899.1000213.

TE

14 25. Fardet L, Kerob D, Rybojad M, et al. Idiopathic systemic capillary leak syndrome: cutaneous

15 involvement can be misleading. Dermatology (Basel). 2004;209(4):291-295.

EP

16 doi:10.1159/000080851.
C

17 26. Dhir V, Arya V, Malav IC, Suryanarayanan BS, Gupta R, Dey AB. Idiopathic systemic capillary leak

18 syndrome (SCLS): case report and systematic review of cases reported in the last 16 years.
AC

19 Intern Med. 2007;46(12):899-904.

20

16
 ACCEPTED MANUSCRIPT

Table 1 General characteristics of 69 patients with monoclonal gammopathy-associated systemic capillary-leak

syndrome.

Parameter n* Value

Patients 69

Women 35 (50.7)

PT
Age at onset§, year 52 ± 12

Age at diagnostic, year 53.5 ± 12

RI
Time from onset to diagnostic, month 5.9 [0.1-20.5]

SC
Monoclonal gammopathy 69 (100)

IgG heavy chain 69 (100)

Kappa light chain¶ 47 (68.1)

U
Lambda light chain¶ 24 (34.8)

Monoclonal protein level at diagnosis, 55

AN 4.4 [2-8]
g/L
M
Trigger 48 (69.5)

Infection¥ 45 (65.2)
D

Hormonal 4 (5.8)
TE

Effort 2 (2.9)

Biological data
EP

Hemoglobin‡, g/dL 20.3 [18.1-22]

Hematocrit‡, % 62 59.6 [55-65.2]

C

Proteinemia†, g/L 60 45 [38-52.7]

Albuminemia†, g/L 59 23.6 [17-28]

AC

Values are expressed as n (%), and continuous variables as mean ± standard deviation or median [interquartile

range].

Abbreviations: IgG, G type immunoglobulin.

*
Number of value available for the 69 patients.
§
Onset was defined as the first attack reported.
 ACCEPTED MANUSCRIPT

¶
Two patients had a biclonal gammopathy.
¥
Including flu-like illness, proven viral and bacterial infections.
‡
Highest value available for each patient during any attack.
†
Lowest value available for each patient during any attack.

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 2 Outcome, preventive treatment and survival of 69 patients with monoclonal gammopathy-associated

systemic capillary-leak syndrome.

Parameter Value

Patients 69

Duration of follow-up, year 5.1 [2.5-9.7]

PT
Number of attacks before diagnosis 2 [1-3]

Severe attacks 1 [1-2]

RI
Chronic evoluGon before diagnosis† 4 (5.8)

SC
Number of attacks after diagnosis 2 [0-6]

Severe attacks 1 [0-2]

Chronic evoluGon aHer diagnosis† 9 (13)

U
Complications

Compartment syndrome
AN 23 (33.3)

Myeloma 5 (7.2)
M
Death 24 (34.8)

Severe attack 20 (83.3)

D

Myeloma 4 (16.7)
TE

*
Patients 65

Preventive treatment 56 (86.2)

EP

More than one preventive treatment 23 (35.4)

Intravenous immunoglobulins 48 (73.8)

C

Duration, year 4.3 [2.3-7.5]

Theophylline 22 (33.8)
AC

Duration, year 2.5 [1.4-3.6]

Terbutaline 22 (33.8)

Duration, year 2.3 [0.7-3.9]

Thalidomide 5 (7.7)

Duration, year 1.4 [0.8-7]

 ACCEPTED MANUSCRIPT

Probability of survival
†
1-year n=60 97%

2-year n=58† 95%

5-year n=35† 78%

10-year n=17† 69%

PT
†
15-year n=6 49%

RI
Values are expressed as n (%), and continuous variables as mean ± standard deviation or median [interquartile

SC
range].
*
Four patients died during their first attack and were not included in the follow-up analysis.
†
Number of patients still at risk of dying, excluding patients whose follow-up is ended or lost.

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

¥
Table 3 Comparison between survivors and non-survivors in 65 patients with monoclonal gammopathy-

associated systemic capillary-leak syndrome.

n* Alive Dead
p
n = 45 n = 20

PT
General characteristics

Women 24 (53.3) 9 (45) 0.6

Age at onset§, year 52.9 ± 13 51.3 ± 11.1 0.9

RI
Age at diagnostic, year 53.8 ± 12.7 52.7 ± 11.3 0.9

SC
Duration of follow-up, year 5.8 [3.4-10.8] 4.5 [2.5-10.3] 0.3

Start of follow-up after year 2000 39 (86.7) 8 (40) <0.0001

Time from onset to diagnostic, month 4.6 [0.1-19.5] 11.1 [1.2-23.9] 0.2

U
Trigger 28 (62.2) 17 (85) 0.08

Monoclonal gammopathy
AN
Kappa light chain 34 (75.5) 11 (55) 0.1
M
Monoclonal component level at 54 0.2
4 [1.5-8.8] 4.1 [1.6-8.8]
diagnosis, g/L
D

Biological data
TE

‡
Hemoglobin , g/dL 20 [18.1-22.4] 20,1 [17.9-21] 0.4
‡
Hematocrit , % 62 59.6 [55-65] 59.3 [57-68] 0.8
EP

Proteinemia†, g/L 60 45 [33.5-52] 41 [39-53] 0.9

Albuminemia†, g/L 59 24 [16.5-28.7] 20.5 [17.9-27.2] 0.8

C

Capillary-leak episodes

Number of attacks before diagnosis 2 [1-3] 2 [1.2-4] 0.1

AC

Severe attacks 1 [1-1.5] 1 [1-2] 0.3

Chronic evolution 2 (4.4) 2 (10) 0.5

Recurrence of attacks after diagnosis 28 (62.2) 19 (95) 0.007

Severe attacks 19 (42.2) 19 (95) <0.0001

Number of attacks after diagnosis 1 [0-8] 3 [1.25-7.25] 0.09

 ACCEPTED MANUSCRIPT

Severe attacks 0 [0-2] 1 [1-6] 0.001

Chronic evolution 4 (8.9) 5 (25) 0.1

Complications

Compartment syndrome 16 (35.5) 4 (20) 0.2

Myeloma 0 (0) 5 (25) 0.002

PT
Treatments

Preventive treatment 43 (95.5) 13 (65) 0.003

RI
More than one preventive treatment 16 (35.5) 7 (35) 1

SC
Intravenous immunoglobulins 40 (88.9) 8 (40) <0.0001

Duration, year 4.1 [2.4-7.5] 4.5 [2.2-9.9] 0.8

Theophylline 14 (31.1) 8 (40) 0.5

U
Duration, year 2.5 [0.9-5.1] 2.3 [0.8-3.2] 0.5

Terbutaline
AN16 (35.5) 6 (30) 0.8

Duration, year 2.5 [0.6-4.7] 1.7 [0.06-2.8] 0.1

M
Thalidomide 2 (4.4) 3 (15) 0.1

Values are expressed as n (%), continuous variables as mean ± standard deviation or median [interquartile
D

range].
TE

*
Number of value available for the 65 patients.
¥
Four patients died on their first attack and were not included in the follow-up analysis.
EP

§
Onset was defined as the first attack reported.
‡
Highest value available for each patient during any attack.
C

†
Lowest value available for each patient during any attack.
AC
 ACCEPTED MANUSCRIPT

Table 4 Cox proportional hazard model univariate and multivariate analysis of all-cause mortality associated
¥
factors in 65 patients with monoclonal gammopathy-associated systemic capillary-leak syndrome.

Univariate Multivariate

HR [95% CI] p value HR [95% CI] p value

PT
Women 0.90 [0.37-2.18] 0.8

*
Age at onset , year 1.01 [0.97-1.05] 0.4 1.01 [0.96-1.06] 0.6

RI
*
Start of follow-up after year 2000 0.53 [0.19-1.45] 0.2 1.31 [0.38-4.47] 0.6

Kappa light chain 0.40 [0.16-0.97] 0.04

SC
Monoclonal protein level at diagnosis > 5g/L 1.31 [0.42-4.11] 0.6

Myeloma* 2.48 [0.87-7.08] 0.09 1.33 [0.46-3.84] 0.6

U
Compartment syndrome 0.81 [0.26-2.54] 0.7

Recurrence of attacks after diagnosis

*
AN
3.38 [0.44-25.7] 0.2

Severe attacks 7.46 [0.98-56.8] 0.052 5.47 [0.67-44.2] 0.1

M
Number of attacks after diagnosis 0.97 [0.91-1.03] 0.3

Severe attacks 1.02 [0.93-1.12] 0.6

D

Preventive treatments 0.18 [0.07-0.49] 0.001

TE

*
Intravenous immunoglobulins 0.22 [0.09-0.54] 0.001 0.27 [0.10-0.70] 0.007

Theophylline 0.80 [0.32-2.01] 0.6

EP

*
Terbutaline 0.48 [0.18-1.29] 0.1 0.35 [0.13-0.96] 0.04

*Factors included in the multivariate analysis.

C

¥
Four patients died on their first attack and were not included in the follow-up analysis.
AC
 ACCEPTED MANUSCRIPT

¥
Table 5 Comparison between patients treated or not with intravenous immunoglobulins in 65 patients with

monoclonal gammopathy-associated systemic capillary-leak syndrome.

IVIg No IVIg
p
n = 48 n = 17

Women 23 (47.9) 10 (58.8) 0.5

PT
Start of follow-up after year 2000 7 (14.6) 11 (64.7) <0.0001

Alive at the end of follow-up 40 (83.3) 5 (29.4) <0.0001

RI
§
Age at onset , year 51.3 ± 13 55.7 ± 9.6 0.2

SC
Age at diagnosis, year 52.2 ± 12.9 57 ± 9 0.2

Duration of follow-up, year 5.4 [3.4-10.2] 4.5 [2.4-11.7] 0.6

Time from onset to diagnostic, month 6 [0.1-21.8] 9.2 [0.6-21.7] 0.5

U
Myeloma 1 (2.1) 4 (23.5) 0.015

Number of attacks before diagnosis

AN
2 [1-3.75] 2 [1-3.5] 0.9

Severe attacks 1 [1-2] 1 [1-2] 0.8

M
Recurrence of attacks after diagnosis 31 (64.6) 16 (94.1) 0.03

Severe attacks 22 (45.8) 16 (94.1) <0.0001

D

Number of attacks after diagnosis 1.5 [0-10.5] 2 [1.5-5] 0.4

TE

Severe attacks 0 [0-2] 1 [1-3.5] 0.01

Values are expressed as n (%), continuous variables as mean ± standard deviation or median [interquartile
EP

range].

Abbreviations; IVIg, intravenous immunoglobulins.

¥
C

Four patients died on their first attack and were not included in the follow-up analysis.
§
Onset was defined as the first attack reported.
AC
 ACCEPTED MANUSCRIPT

¥
Table 6 Comparison between patients treated or not with terbutaline in 65 patients with monoclonal

gammopathy-associated systemic capillary-leak syndrome.

Terbutaline No Terbutaline
p
n = 22 n = 43

Women 14 (63.6) 19 (44.2) 0.2

PT
Start of follow-up after year 2000* 12 (54.5) 35 (81.4) 0.04

Alive at the end of follow-up 16 (72.7) 29 (67.4) 0.8

RI
§
Age at onset , year 52.7 ± 11.4 52.3 ± 12.9 0.9

SC
Age at diagnosis, year 53.8 ± 10.6 53.3 ± 13.1 0.8

Duration of follow-up, year 7.3 [4.7-12.8] 4.5 [2.5-8.1] 0.014

Time from onset to diagnostic, month 6.9 [0.5-21] 8.1 [0.1-22.3] 0.5

U
Myeloma 1 (4.5) 4 (9.3) 0.6

Number of attacks before diagnosis

AN
2 [1-3] 2 [1-4] 0.5

Severe attacks 1 [1-2] 1 [1-2] 0.2

M
Recurrence of attacks after diagnosis 20 (90.9) 27 (62.8) 0.02

Severe attacks 17 (77.3) 21 (48.8) 0.035

D

Number of attacks after diagnosis 5 [2-14] 1 [0-5] 0.003

TE

Severe attacks 2 [0.75-5] 0 [0-1] 0.006

Values are expressed as n (%), continuous variables as mean ± standard deviation or median [interquartile
EP

range].
¥
Four patients died on their first attack and were not included in the follow-up analysis.
§
C

Onset was defined as the first attack reported.

AC
 ACCEPTED MANUSCRIPT

¥
Table 7 Comparison according the occurrence of a severe attack after diagnosis in 65 patients with

monoclonal gammopathy-associated systemic capillary-leak syndrome.

Severe Recurrence No Severe Recurrence

p
n = 38 n = 27

Women 19 (50) 14 (51.9) 1

PT
Start of follow-up after year 2000 21 (55.3) 26 (96.3) <0.0001

Alive at the end of follow-up 19 (50) 26 (96.3) <0.0001

RI
§
Age at onset , year 50.7 ± 11.3 54.8 ± 13.1 0.2

SC
Age at diagnosis, year 52.1 ± 11.3 55.5 ± 13.4 0.3

Duration of follow-up, year 6.65 [3.4-12.7] 4.5 [2.3-7.1] 0.01

Time from onset to diagnostic, month 8.9 [0.5-22.8] 0.6 [0.1-16.5] 0.1

U
Monoclonal component level at
diagnosis > 5g/L AN
17 (44.7) 7 (25.9) 0.006

Myeloma 4 (10.5) 1 (3.7) 0.4

M
Number of attacks before diagnosis 2 [1-4] 2 [1-3] 0.1

Severe attacks 1 [1-2] 1 [1-2] 0.4

D

Attacks after diagnosis 38 (100) 9 (33.3) <0.0001

TE

Severe attacks 38 (100) 0 (0) na

Number of attacks after diagnosis 5 [2-14] 0 [0-1] na

Severe attacks 2 [1-5] 0 [0-0] na

EP

Preventive treatments 30 (78.9) 26 (96.3) 0.02

Intravenous immunoglobulins 22 (57.9) 26 (96.3) <0.0001

C

Theophylline 19 (50) 3 (11.1) 0.001

AC

Terbutaline 17 (44.7) 5 (18.5) 0.035

Thalidomide 5 (13.2) 0 (0) 0.07

Values are expressed as n (%), continuous variables as mean ± standard deviation or median [interquartile

range].
¥
Four patients died on their first attack and were not included in the follow-up analysis.
§
Onset was defined as the first attack reported.
 ACCEPTED MANUSCRIPT

Figure legends.

Figure 1. Treatment administration distribution in 65 patients with monoclonal gammopathy-

associated systemic capillary-leak syndrome.

PT
Figure 2. Kaplan-Meier curve of cumulative probability of survival in 65 patients with monoclonal

gammopathy-associated systemic capillary-leak syndrome.

RI
Figure 3. Kaplan-Meier curve of cumulative probability of survival in 65 patients with monoclonal

SC
gammopathy-associated systemic capillary-leak syndrome with or without preventive treatment with

intravenous immunoglobulin.

U
Abbreviations; IVIg +, patients treated with intravenous immunoglobulin; IVIg -, patients not treated
AN
with intravenous immunoglobulin. Probabilities of survival were compared using the log rank test.
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Figure 1.

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Figure 2.

PT
RI
U SC
AN
Duration of follow-up (years) 0 1 2 4 6 8 10
M
Patients at risk 65 60 58 43 29 22 17
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Figure 3.

PT
RI
U SC
AN
M

Duration of follow-up (years) 0 1 2 4 6 8 10

D

IvIg + 48 45 44 34 21 15 12
Patients at risk
TE

IvIg - 17 15 14 8 8 5 4
C EP
AC
 ACCEPTED MANUSCRIPT

1 Clinical significance

2 Five- and ten-year survival rates in 65 patients with monoclonal gammopathy-associated

3 systemic capillary-leak syndrome were 78% and 69%, respectively.

PT
5 Intravenous immunoglobulins improve survival of patients with monoclonal gammopathy-

6 associated systemic capillary-leak syndrome.

RI
U SC
AN
M
D
TE
C EP
AC