Secondary Logo

Journal Logo

Whole-Exome Sequencing of Adult and Pediatric Cohorts of the Rare Vascular Disorder Systemic Capillary Leak Syndrome

Pierce, Richard*; Ji, Weizhen*; Chan, Eunice C.; Xie, Zhihui; Long, Lauren M.; Khokha, Mustafa*,‡; Lakhani, Saquib*; Druey, Kirk M.

doi: 10.1097/SHK.0000000000001254
Clinical Science Aspects
Editor's Choice

Objective: Systemic capillary leak syndrome (SCLS) is a rare disorder that presents with episodes of hypovolemic shock. The extent to which genetic abnormalities contribute to SCLS is unknown. We identified pediatric and adult cohorts with characteristic clinical courses. We sought to describe the clinical characteristics of both cohorts, identify a possible genetic contribution to SCLS, and demonstrate that whole-exome sequencing (WES) may be conducted by critical care providers.

Design: Prospective observational study of WES of nine adult and eight pediatric SCLS patients and available unaffected first-degree relatives.

Setting: Tertiary children's hospitals and referral research laboratory.

Patients: Children and adults with SCLS.

Interventions: None.

Measurements: Patients and available first-degree relatives underwent WES. Data were analyzed for rare homozygous, biallelic, de novo, and heterozygous variants with allelic enrichment and metabolic pathway analyses.

Main Results: Children with SCLS presented at a younger age with episodes similar to those experienced by adults. All patients and available relatives underwent satisfactory WES. No overlapping gene variants or metabolic pathways were identified across all SCLS patients. Multiple candidate genes with homozygous or biallelic mutations were identified in individual subjects with SCLS. There was no significant enrichment of genes with rare heterozygous variants.

Conclusions: The clinical characteristics of children and adults with SCLS are similar. We did not identify a uniform germline exomic genetic etiology for SCLS. WES identified several candidate genes in individual patients for future research. WES is a viable way for critical care providers to investigate the etiology of diseases with presumed genetic contributions.

*Pediatric Genomic Discovery Program, Department of Pediatrics, Yale University, New Haven, Connecticut

Molecular Signal Transduction Section, Laboratory of Allergic Diseases, National Institute of Health, Institute of Allergy and Infectious Disease, Bethesda, Maryland

Department of Genetics, Yale University, New Haven, Connecticut

Address reprint requests to Richard Pierce, MD, MS, 333 Cedar Street, PO Box 208064, New Haven, CT 06520. E-mail:

Received 22 June, 2018

Revised 16 July, 2018

Accepted 15 August, 2018

Supplemental digital content is available for this article. Direct URL citation appears in the printed text and is provided in the HTML and PDF versions of this article on the journal's Web site (

This work is supported by the National Institutes of Health grants (R33HL120783 to MKK and T32HD068201 to RP) and in part by the Intramural Research Program, NIAID/NIH. M.K. Khokha is a Mallinckrodt Scholar. The Pediatric Genomics Discovery Program is supported by Yale New Haven Hospital, Yale Medicine, and the generosity of Sara and Jeffery R. Buell.

The authors report no conflicts of interest.

Back to Top | Article Outline


The genetic contribution to many illnesses seen in adult and pediatric intensive care units (ICUs) is increasingly recognized. Many of these syndromes are associated with a high morbidity and mortality (1). Prompt diagnosis and characterization of these disorders in the short term may assist with appropriate medical management as well as long-term family counseling (2). In addition, identification of genetic defects, within their clinical context, can greatly aid scientific investigations into these complicated diseases. Such clinician-initiated investigations will ultimately improve our understanding of pathophysiology as well as identify potential therapeutic targets.

As an example of this approach of blending clinical assessment and genetic investigation, we examined systemic capillary leak syndrome (SCLS). First described by Clarkson in 1960, this rare and distinctive condition (3) is characterized by recurrent, transient, and spontaneous episodes of massive shifts of intravascular fluids into peripheral tissues, resulting in intractable hypovolemic shock (4). Both children and adults with SCLS present in hypovolemic shock with hemoconcentration and hypoproteinemia due to profound vascular leak. Episodes may vary in severity from mild eyelid or scrotal swelling to profound systemic edema with organ dysfunction and cardiac arrest. Upon presentation to the emergency department or an ICU, most SCLS patients are treated with massive infusions of intravenous saline, which are promptly “third-spaced,” leading to anasarca, particularly of the face, trunk, and extremities, with relative sparing of the lungs. As a result, compartment syndrome may develop in the peripheral limbs, frequently necessitating fasciotomies to prevent muscle ischemia or even loss of limbs. Other complications arise from organ parenchymal edema, such as multiple organ dysfunction, and from perturbed blood flow, such as disseminated intravascular coagulation. Symptoms typically resolve spontaneously after 48 to 96 h without specific treatment other than intense hemodynamic support and treatment of complications such as compartment syndromes.

Although the presenting signs and symptoms of SCLS may overlap with more common diseases associated with vascular leak including sepsis, anaphylaxis, or acute infections with filovirus family, such as Ebola or Marburg hemorrhagic fevers (5, 6), SCLS has unique features which are simply not observed in other vascular leak syndromes leading to hypovolemic shock. Simultaneously occurring hypotension, hemoconcentration, hypoalbuminemia, and anasarca with sparing of the lungs are quite unique to SCLS. Hemoconcentration (elevated hemoglobin and hematocrit) is often so severe that it may prompt misdiagnosis as polycythemia vera and treated inappropriately with phlebotomy (4). Although SCLS flares are frequently preceded by infections (7), these are typically mild viral upper respiratory or gastrointestinal infection, and blood cultures are negative. There are typically no signs of excessive or uncontrolled inflammation, as seen in sepsis. Unlike anaphylaxis, allergic triggers are not apparent, and hypotensive shock in SCLS is refractory to vasoconstrictors such as epinephrine. In contrast to hemorrhagic fever syndromes, bleeding into tissues is not observed during acute SCLS episodes. Finally, a feature of adult SCLS not found in other vascular leak syndromes is a monoclonal gammopathy of unknown significance (MGUS), which is present in up to 80% of adult patients with SCLS, but has not been observed in children with SCLS.

The etiology of SCLS is unknown. Current hypotheses include abnormal endothelial cell response to normal stimulation, or normal endothelial cell response to dysregulated or excessive systemic signaling or inflammation. Both hypotheses presuppose exaggerated or abnormal cellular signaling, either in vascular or inflammatory cells, which could be explained by germline mutations in genes that participate in such signaling. The presence of MGUS is not considered essential for the diagnosis of SCLS because it is not universally present and the pathological role for the MGUS “paraprotein” IgG, if any, has not been established (8). No disease pedigrees can be identified, suggesting subtle genetic abnormalities, potentially involving redundant vascular permeability or proinflammatory signaling pathways. A better understanding of SCLS could provide insight into the pathophysiology of vascular leak seen in other shock-associated syndromes such as sepsis, burns, cardiac arrest, or multiple dysfunction syndrome (MODS).

Utilizing whole-exome sequencing (WES), we recently identified the etiology of capillary leak in a child (patient 10 below) presenting with many of the clinical features of SCLS who died from complications of hypovolemic shock and circulatory collapse (9). Using postmortem tissue samples from this patient, we identified a novel single gene mutation in a protein associated with endothelial permeability signaling pathways, p190BRhoGAP. This protein is a GTPase accelerating protein (GAP) which, along with guanine nucleotide exchange factors (GEFs), regulates the GTP-bound state and therefore activity of RhoGTPases (Fig. 1). Although a handful of RhoGTPases are known to modulate endothelial cell permeability, attachment, and intra- and intercellular tensile forces (10), more than 60 GAPs and GEFs and 20 RhoGTPases exist, most with unassigned functions. Mutations in previously uncharacterized regulatory proteins related to RhoGTPase signaling could explain the lack of a previously identified genetic susceptibility of SCLS. Based on the results obtained from this single case, we performed WES on pediatric and adult patients with SCLS to investigate whether or not defects in RhoGTPase signaling or other permeability pathways represent a common pathophysiologic feature of SCLS.

Fig. 1

Fig. 1

In this study, we report the characteristics of an adult and pediatric cohort with clinically diagnosed with SCLS. In addition, WES results from a pediatric cohort of seven children with SCLS, their unaffected first-degree relatives, and a pediatric patient whose first-degree relatives were not available. Our first objective was to identify a genetic contribution to SCLS. Secondary objectives were to report the clinical characteristics of both SCLS cohorts. Finally, we sought to demonstrate how WES may be obtained and analyzed in critically ill patients. It is our hope to show the utility and feasibility of this combined clinical and genetic approach to other clinicians.

Back to Top | Article Outline



Patients and willing first-degree relatives were recruited to the NIH through a National Institute of Allergy and Infectious Disease (NIAID) Institutional Review Board (IRB)-approved protocol (I-0184). All patients signed informed consent. Both children and adults were diagnosed with SCLS by the senior author based upon having experienced at least one episode conforming to published clinical criteria (11, 12) and the absence of underlying causes for acute hypotension, hemoconcentration, hypoalbuminemia, and peripheral edema. Adult SCLS patients, designated 1 through 9, were sequenced at the National Institutes of Health (NIH), NIAID. Pediatric SCLS patients, designated 10 through 17, and their available relatives, were sequenced through the Pediatric Genomics Discovery Program at Yale University School of Medicine.

Back to Top | Article Outline

Whole-exome sequencing

De-identified genomic DNA samples were provided to Yale under its IRB-approved protocol with waiver of consent. Exome sequencing of adult patients was conducted at NIH using genomic DNA isolated from peripheral blood mononuclear cells and performed by Agilent exome enrichment followed by Illumina Hiseq2000 sequencer, as previously described (13). Exome sequencing of pediatric patients and their family members was conducted at the Yale Center for Genomic Analysis through the Pediatric Genomics Discovery Program (PGDP,, a program that is freely available to coordinate sequencing, analysis, and additional testing with pediatric critical care researchers caring for children with diseases of suspected genetic etiology. Capture was performed using the NimbleGen 2.1 exome capture reagent or SeqxCap EZ MedExome Target Enrichment Kit (Roche) followed by Illumina DNA sequencing (HiSeq2500), as previously described (14). Sequence reads from both cohorts were converted to FASTQ format and were aligned to the reference human genome (hg19). Genetic variants were called by GATK, and they were annotated by ANNOVAR and a custom pipeline that includes allele frequencies, OMIM and ClinVAR citations, and numerous in silico attributes (15). To determine the ethnicity of each sample by principal component analysis, we used the EIGENSTRA (16) software to analyze tagSNPs that are best representative particular haplotypes in the genome from all the samples and HapMap subjects, as previously described (17).

Back to Top | Article Outline


Clinical data

The median age of adult patients at diagnosis was 48 years (IQR). They had a median of three severe episodes (requiring intensive care) since diagnosis. In the adult cohort, a single patient was deceased, and four of the eight living patients were receiving ongoing intravenous immunoglobulin therapy. More than half of the adult SCLS patients had an associated peripheral neuropathy or myopathy complicating compartment syndromes that accompany the severe edema of acute SCLS flares and all had MGUS, a prominent feature of SCLS in adults (Table 1). In the pediatric cohort, the median age of first episode was seven years (IQR) with a median of two severe episodes since their original diagnosis (Table 2). One of the pediatric patients was deceased and all living patients were receiving immunoglobulin therapy. Six of the eight pediatric patients had episodes linked to viral infectious triggers, a feature shared with less than half of adult SCLC patients (18, 19). The most common serious complications of SCLS in children were pleural or pericardial effusions and no pediatric patients had MGUS.

Table 1

Table 1

Table 2

Table 2

Back to Top | Article Outline

Whole-exome sequencing metrics

The whole exomes of the seven pediatric SCLS patients and their unaffected first-degree relatives (in total 26 samples, Supplemental Figure 1A, were sequenced to a mean depth of 76.5 independent reads per targeted base across all the samples. Across all samples, 97.1% of all targeted bases had more than eight independent reads. The nine adult SCLS patients were sequenced previously to a mean depth of 35.8 reads. The mean of 83.7% of targeted bases, across the nine samples, had more than 8X coverage. This depth provides enough-confidence calling of homozygous and heterozygous variants across the exomes (Supplementary Tables 1 and 2,

Principal component analysis (PCA) from WES genotypes showed that the majority of patients were of European ancestry with one patient of each African ancestry and Mexican ancestry (Supplemental Figure 2B, This result agrees with the patients’ self-identified ethnicities. The minor allele frequency used to analyze each sample was subsequently adjusted to account for ethnicity.

Back to Top | Article Outline

De novo variant analysis

No SCLS patients in either cohort had any family history of the disease, rendering traditional gene mapping strategies in extended families impossible. Therefore, we analyzed parent–child trios, a validated approach for gene discovery, with a focus on rare recessive and de novo variants (20). There were five such trios available in the current study. We identified only one de novo coding variant (c.A5765G, p.Q1922R) in the DNAH11 gene from one family (Table 3). This variant was novel as it was not identified in 123,126 exomes in the gnomAD database, and the encoded amino acid was highly conserved across human evolution. DNAH11 encodes Dynein Axonemal Heavy Polypeptide 11, a protein involved in microtubule-dependent motor ATPase movement of respiratory cilia; mutations have been associated with primary ciliary dyskinesia (21). However, DNAH11 is also expressed in endothelial cell lines, and microtubule function has essential roles in vascular barrier function (22).

Table 3

Table 3

Back to Top | Article Outline

Recessive inheritance analysis

Next, we searched for rare homozygous or compound heterozygous disease-causing nucleotide variants that may follow recessive inheritance. Among our cohort, there were seven patients with first-degree relatives available for sequence comparisons. Parental study allows for ready determination of whether any particular allele is inherited from the maternal or paternal side. Frequently, sequence analysis of only one parent may permit such a determination when sequence from the other parent is not available. If only one allele is present in the available parent, the other allele either must have been inherited from the other parent or occurred de novo. Using this approach, we could distinguish between compound heterozygous and biallelic variants if they are inherited in trans. If the inheritance of the biallelic variants is not distinguishable due to lack of parent genetic testing, and they were present in the 1,000 genomes dataset with strong linkage disequilibrium (D’ > 0.9) demonstrated within a population, they were excluded. Because a recombination between them is unlikely to be happened in the genome. We identified multiple compound heterozygous (Table 4), biallelic heterozygous (Supplementary Table 3,, and homozygous or X-linked hemizygous (Supplementary Table 4, variants that warranted investigation as possibly contributing to the pathogenesis of SCLS. We included variants with a relatively high allele frequency (≤0.5% subpopulation frequency) and applied liberal criteria to assess the impact of nucleotide variations on protein function and disease-causing potential (combined annotation dependent depletion [CADD] score ≥15) to maximize the possibility for discovery, which may identify a few false positive target genes. Despite these relatively low thresholds, we did not identify rare gene variants common to all patients, nor were there mutations in genes that were common to multiple patients but containing distinct mutations in individual patients.

Table 4

Table 4

Back to Top | Article Outline

Rare heterozygous variant analysis

Finally, we sought rare heterozygous variants in the nine unrelated adult subjects. We used a dominant inheritance model to identify novel or highly rare alleles (MAF≤0.01%) and focused on variations predicted to be highly deleterious (CADD score ≥20) or lead protein-truncation. We also used gene-based collapsing analysis of these variants (15) for discovery of genes containing potentially pathogenic mutations. Collapsing analysis identifies variations of potential pathologic importance in all relatives, thereby producing a large list of potentially implicated genes, which much is analyzed in the appropriate clinical and biological pathway context. To maximize the impact of the identified gene targets, we performed extensive quality control and screened each aligned read to avoid any false positive variants that could give spurious findings. The analysis of very rare heterozygous protein-truncating variants yielded 46 genes of interest, which are listed in Supplementary Table 5 ( No common genes or metabolic pathways with rare protein-truncating heterozygous variants were identified. The gene-based analyses, which collapse all rare loss of function (LOF) and damaging variants on a gene level, did not provide any statistically significant enrichment for a disease-associated gene. This analysis identified 10 genes of potential interest and are listed in Supplementary Table 6 (

Back to Top | Article Outline


In this report, we detail the clinical characteristics and WES genetic analysis from pediatric and adult SCLS cohorts. This investigation represents the most comprehensive genetic analysis to date exploring a potential underlying genetic etiology of SCLS. Although no common genes or pathways were detected across either the adult or pediatric cohorts, in-depth analysis identified multiple single nucleotide variations that may have relevance to individual patients. These results suggest that the SCLS phenotype is not explained by germline exomic mutations either in a single gene or known pathway. Our results do not definitively exclude genetic variation as a contributing etiology to SCLS. Other mechanisms of genetic influence may be considered, such as locus heterogeneity, when mutations in several different genes can lead to a similar clinical phenotype. In complex biological processes such as vascular permeability, a large number of genes may interact to produce a common phenotype. This hypothesis could be tested by further basic science research into rare variants in various genes that were identified in individual patients in our cohort. Furthermore, SCLS could also be a result of genetic changes in noncoding regions, such as variants in intronic sequences or epigenetic modifications. Evaluation of this possibility would require additional approaches such as whole-genome sequencing or DNA methylation analyses. A third alternative is that SCLS could result from complex interactions of variants of two or more genes. Our current study does not provide support for or against these more complex genetic mechanisms of genetic regulation in SCLS.

Nongenetic etiologies for SCLS must also be considered. For example, the paraprotein associated with MGUS may act through an as-yet undescribed mechanism to induce leak in adult patients, although this would not explain pediatric patients with SCLS. Other possibilities include occult infection or subclinical insult that alters the systemic permeability or inflammatory response. Possibilities include viral infections that alter VEGF, angiopoietin-2, tumor necrosis factor or IL-2 signaling (8), or an unknown process that promotes endothelial apoptosis (23). Evidence for such competing hypothesis is scarce; however, our results suggest that investigations directed at novel hypotheses may be worthwhile.

In pursuing this study, we also show the feasibility and applicability of WES in investigating the etiology of disease with presume genetic contribution in the ICU setting. Historically, when caregivers suspected a genetic component to their patient's illness, several obstacles delay the rapid identification of monogenic defects contributing to disease. Specimens may be difficult to obtain from pediatric or acutely ill adult patients. In addition, even with timely acquisition and analysis of sequencing, interpretation of these tests may be attempted by researchers with little understanding of the clinical syndrome in question. Critical care clinician researchers and their trainees, on the contrary, are well positioned to combine clinical knowledge and data with genetic analysis which may lead to meaningful discoveries (24). Such innovation is already underway in neonatal intensive care units where genomic technologies are being increasingly used (25). As we have demonstrated here, such patient-focused research efforts have the potential to improve our understanding of how specific genetic abnormalities contribute to disease pathogenesis. Although no unifying genetic etiology of SCLS was identified, we uncovered a number of noteworthy findings that may impact the care of current and future SCLS patients.

Back to Top | Article Outline

Clinical characteristics

The clinical characteristics reported in these cohorts are consistent with those previously reported (4, 26). Although a consistent feature of pediatric SCLS is its linkage to antecedent infections, infectious triggers are more variably (50%–75%) identified in adults (7). As previously reported, MGUS associated with adult SCLS patients was not observed in our pediatric cohort (26). In addition, the overall number of severe episodes and mortality in both cohorts are lower than previously reported, possibly due to the recent advance of immunoglobulin therapy to prevent serious episodes (12, 27). The high mortality in those presenting with undiagnosed disease supports his theory. In short, the adults and children studied are highly clinically representative of SCLS cases reported in the medical literature and thus an excellent cohort for WES analysis.

Back to Top | Article Outline

De novo variant analysis

Sequencing of trios is a powerful tool to detect the presence nucleotide variants present in probands absent in the parents (Table 3). Such variations are often highly significant for disease processes, although there are natural variations in genomic DNA that may have no impact on protein functions. Our analysis found only one individual with a single de novo variation in the gene DNAH11, which encodes dynein axonemal heavy chain 11, a protein component of cilia. It is mainly expressed in ciliated cells in the fallopian tubes and respiratory tract, as well as in parathyroid and adrenal glands (28). Mutations of the gene are associated with Ciliary Dyskinesia and situs inversus, but there are no reports of vascular leak-related symptoms in these disorders (29). Although DNAH11 variants are associated with lethal respiratory and cardiac lesions in mice (30), there is no mechanistically plausible link of DNAH11 to SCLS. WES may detect small numbers (less than three) of normal de novo coding variants in any given patient, raising the likelihood that this finding is simply a rare normal variant and unlikely to explain the patient's SCLS.

Back to Top | Article Outline

Recessive and compound heterozygous inheritance analysis

Inheritance of two defective alleles results in autosomal recessive disorders. We identified 21 genes with rare single nucleotide variations that were likely to adversely affect protein function as assessed by in silico prediction algorithms. A single adult patient had biallelic mutations in DOCK6, although compound heterozygous status is uncertain due to lack of family inheritance information. DOCK6 encodes Zir1, a GEF for the RhoGTPases Cdc42 and Rac1, both of which are known to enhance vascular barrier function through stabilization of tight junctions (30). As a GEF, Zir1 would function as a molecular “on” switch for Cdc42 and Rac1 (Fig. 1). A defect in the “on” switch for a RhoGTPase that acts to fortify vascular barriers may very well produce the signs and symptoms of excessive vascular leak observed in SCLS. Some mutations in DOCK6 in humans have been associated with congenital finger and toe abnormalities in Adams-Oliver Syndrome (31). Finally, although it is widely expressed in multiple tissues, DOCK6 exhibits highest expression levels in endothelial cell lines, making it a potentially attractive target. We are currently planning further studies into its role in endothelial permeability pathways.

Multiple other genes of potential interest for SCLS pathogenesis were identified as noted in Table 4 and Supplemental Table 3 ( Although numerous, these results must be interpreted in the specific clinical setting, highlighting the importance of pediatric critical care providers engaging in this research. Genes that are involved in cell signaling include PIGT, SH3KBP1, MST1R and PELP1. PIGT encodes a glycosylphosphatidylinositol (GPI)-anchor that cleaves membrane bound molecules. Mutations in this gene have been associated with developmental delay (32). Other genes identified include CACNAC1C (a voltage gated calcium channel), AIFM3 (a mitochondrial apoptosis sensor), DNAH17 (a cilia component), F11 (coagulation factor XI), PASD1, ZNF407 (both nuclear binding proteins), PAGE3, SRP68, NPC1L1, TECPR1 (all protein binding genes), TANC2, SRPX, PAPLN, and PXDN (all matrix proteins). To our knowledge, none of these gene products has been implicated in endothelial dysfunction. This provides us with the unique opportunity to potentially uncover dysregulated expression patterns and/or unexpected functions related to vascular barriers in one or more of these candidates. Of note, although the functional significance remains unclear, CACNAC1C was identified as enriched in SCLS patients using genome wide association studies (13).

Back to Top | Article Outline

Rare heterozygous protein-truncating and deleterious missense variant analysis

Due to the extremely low minor allele frequencies, detection of individual rare causal variants is difficult. To increase the power for detection, rare variant burden testing usually requires aggregation of the heterozygous rare LOF and deleterious variants (CADD ≥20) into a gene or a pathway as a unit for statistically significant enrichment. We did not detect any common pathways shared by rare protein-truncating, deleterious variants after adjustment by genetic constraint or size of the gene; however, we did identify individual genes of interest (Supplementary Table 4, A good example is JUP, the gene encoding plakoglobin. Expressed at endothelial cell junctions, plakoglobin prevents barrier dysfunction induced by permeability mediators such as vascular endothelial growth factor A (VEGFA) by interacting with junctional proteins such as the receptor tyrosine phosphatase, VE-PTP, and VE-cadherin (33). As serum levels of VEGFA are increased in patients with SCLS specifically around disease flares, impaired plakoglobin function could induce and/or magnify the clinical symptoms of vascular leakage in these patients (8, 34). JUP is also highly expressed in the cytoplasm and membrane of epithelial cells, especially in the esophagus and skin, neither of which are specifically more affected in SCLS. Mutations of JUP reported thus far have resulted in primarily in dermatologic disease, although cardiac disease has also been described (35).

Our study has several limitations. Because SCLS is a highly rare disease, we had limited patient enrollment. We sequenced eight children and nine adults with SCLS, and thus our study is underpowered due to the extreme rarity of the disease. Increased patient enrollment will greatly assist in discovery, especially for rare heterozygous loss of function variants. Three of our pediatric patients and all nine of our adult patients did not have both parents sequenced, limiting our de novo analysis.

Back to Top | Article Outline

Future directions

We were unable to identify variants in a gene or pathway common to all patients via WES. However, our previous work with a single SCLS patient (9) suggests that genetic defects may still contribute to exaggerated vascular permeability signaling with clinical consequences that match the signs and symptoms of SCLS. Given the lack of a single identified germline exomic etiology, other possible mechanisms include somatic (nongermline) mutations of low frequency that are below the threshold of detection by WES, mutations in noncoding DNA regions, unknown epigenetic changes, polygenic interactions, or some high polymorphic or repetitive regions not optimized by the current method detection. Furthermore, the extraordinarily complex Rho-GAP-GEF signaling network contains many unknown regulators, and it is possible that some of the identified variants affect vascular permeability through pathways that have yet to be identified.

The next steps in the analysis of these data are 2-fold. First, we will continue to locate and enroll new SCLS patients to perform WES to increase the size of our clinical and genetic dataset. Second, we will continue our in vitro investigations of the high priority genes such as DOCK6. Should whole-genome sequencing become more readily available, we may perform this technique on all collected patients to investigate nonprotein coding abnormalities. Finally, we have shown the feasibility of using WES to diagnose critically ill ICU patients. WES is readily available through programs like the Pediatric Genomics Discovery Program at Yale School of Medicine, and should be considered by intensivists and their trainees caring for critically ill patients with clinical symptoms and syndrome suspicious for underlying genetic defects.

Back to Top | Article Outline


SCLS is a rare and potentially lethal clinical phenomenon with a unique yet consistent constellation of signs and symptoms in both pediatric and adult patients. Our study failed to identify a uniform exomic genetic etiology or implicate a single molecular pathway as an underlying cause of SCLS. Other etiologies of SCLS should be considered in future studies, such as nonexomic genetic abnormalities, polygenic interactions, dysregulated inflammatory or vascular permeability signaling, infection, or MGUS. Our WES investigation did identify several candidate genes in individual patients, thereby creating a viable platform for the detection of clinically relevant protein mutations for future in vivo analysis. Finally, through our program's ongoing investigations, we demonstrated that WES may have direct benefits for critically ill patients.

Back to Top | Article Outline


The authors thank the patients and families, who are the inspiration for this study, and also thank the Yale Center for Genome Analysis (YCGA) for exome sequencing support including the assistance by James Knight for sequence alignment algorithms.

Back to Top | Article Outline


1. Cunniff C, Carmack JL, Kirby RS, Fiser DH. Contribution of heritable disorders to mortality in the pediatric intensive care unit. Pediatrics 95:678, 1995.
2. Meng L, Pammi M, Saronwala A, Magoulas P, Ghazi AR, Vetrini F, Zhang J, He W, Dharmadhikari AV, Qu C, et al. Use of exome sequencing for infants in intensive care units: ascertainment of severe single-gene disorders and effect on medical management. JAMA Pediatr 171 12:e173438, 2017.
3. Clarkson B, Thompson D, Horwith M, Luckey EH. Cyclical edema and shock due to increased capillary permeability. Am J Med 29 2:193–216, 1960.
4. Druey KM, Parikh SM. Idiopathic systemic capillary leak syndrome (Clarkson disease). J Allergy Clin Immunol 140 3:663–670, 2017.
5. Martines RB, Ng DL, Greer PW, Rollin PE, Zaki SR. Tissue and cellular tropism, pathology and pathogenesis of Ebola and Marburg viruses. J Pathol 235 2:153–174, 2015.
6. Xie Z, Chen WS, Yin Y, Chan EC, Terai K, Long LM, Myers TG, Dudek AZ, Druey KM. Adrenomedullin surges are linked to acute episodes of the systemic capillary leak syndrome (Clarkson disease). J Leukoc Biol 103 4:749–759, 2018.
7. Eo TS, Chun KJ, Hong SJ, Kim JY, Lee IR, Lee KH, Eisenhut M, Kronbichler A, Shin JI. Clinical presentation, management, and prognostic factors of idiopathic systemic capillary leak syndrome: a systematic review. J Allergy Immunl Clin Pract 6 2:609–618, 2018.
8. Xie Z, Ghosh CC, Patel R, Iwaki S, Gaskins D, Nelson C, Jones N, Greipp PR, Parikh SM, Druey KM. Vascular endothelial hyperpermeability induces the clinical symptoms of Clarkson disease (the systemic capillary leak syndrome). Blood 119 18:4321–4332, 2012.
9. Pierce RW, Merola J, Lavik JP, Kluger MS, Huttner A, Khokha MK, Pober JS. A p190BRhoGAP mutation and prolonged RhoB activation in fatal systemic capillary leak syndrome. J Exp Med 214 12:3497–3505, 2017.
10. Cerutti C, Ridley AJ. Endothelial cell-cell adhesion and signaling. Exp Cell Res 358 1:31–38, 2017.
11. Druey KM, Greipp PR. Narrative review: the systemic capillary leak syndrome. Ann Intern Med 153:90–98, 2010.
12. Pineton de Chambrun M, Gousseff M, Mauhin W, Lega JC, Lambert M, Riviere S, Dossier A, Ruivard M, Lhote F, Blaison FG, et al. EureClark study: intravenous immunoglobulins improve survival in monoclonal gammopathy-associated systemic capillary-leak syndrome. Am J Med 2017; 130 10: 1219.e19–1219.e27, 2017.
13. Xie Z, Nagarajan V, Sturdevant DE, Iwaki S, Chan E, Wisch L, Young M, Nelson CM, Porcella SF, Druey KM. Genome-wide SNP analysis of the systemic capillary leak syndrome (Clarkson disease). Rare Dis 2013; 1 1: pii: e27445, 2013.
14. Romberg N, Al Moussawi K, Nelson-Williams C, Stiegler AL, Loring E, Choi M, Overton J, Meffre E, Khokha MK, Huttner AJ, et al. Mutation of NLRC4 causes a syndrome of enterocolitis and autoinflammation. Nat Genet 46 10:1135–1139, 2014.
15. Seidelmann SB, Smith E, Subrahmanyan L, Dykas D, Abou Ziki MD, Azari B, Hannah-Shmouni F, Jiang Y, Akar JG, Marieb M, et al. Application of whole exome sequencing in the clinical diagnosis and management of inherited cardiovascular diseases in adults. Circ Cardiovasc Genet 2017; 10 1: pii:e001573.
16. Price AL, Patterson NJ, Plenge PM, Weinblatt ME, Shadick NA, Reich D. Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet 38 8:904–909, 2006.
17. Lemaire M, Fremeaux-Bacchi V, Schaefer F, Choi M, Tang WH, Le Quintrec M, Fakhouri F, Taque S, Nobili F, Martinez F, et al. Recessive mutations in DGKE cause atypical hemolytic-uremic syndrome. Nat Genet 45 5:531–536, 2013.
18. Gousseff M, Arnaud L, Lambert M. The systemic capillary leak syndrome: a case series of 28 patients from a European registry. Ann Intern Med 154 7:464–471, 2011.
19. Eo TS, Chun KJ, Hong SJ, Kim JY, Lee IR, Lee KH, Eisenhut M, Kronbichler A, Shin JI. Clinical presentation, management, and prognostic factors of idiopathic systemic capillary leak syndrome: a systematic review. J Allergy Clin Immunol Pract 6 2:609–618, 2018.
20. Rios JJ, Delgado MR. Using whole-exome sequencing to identify variants inherited from mosaic parents. Eur J Hum Genet 23 4:547–550, 2015.
21. Lai M, Pifferi M, Bush A, Piras M, Michelucci A, Di Cicco M, del Grosso A, Quaranta P, Cursi C, Tantillo E, et al. Gene editing of DNAH11 restores normal cilia motility in primary ciliary dyskinesia. J Med Genet 53 4:242–249, 2016.
22. Karki P, Birukova AA. Microtubules-associated Rac regulation of endothelial barrier: a role of Asef in acute lung injury. J Investig Med 65 8:1089–1092, 2017.
23. Assaly R, Olson D, Hammersley J, Fan PS, Liu J, Shapiro JI, Kahaleh MB. Initial evidence of endothelial cell apoptosis as a mechanism of systemic capillary leak syndrome. Chest 120 4:1301–1308, 2001.
24. Bertier G, Sénécal K, Borry P, Vears DF. Unsolved challenges in pediatric whole-exome sequencing: a literature analysis. Crit Rev Clin Lab Sci 54 2:134–142, 2017.
25. Smith LD, Willig LK, Kingsmore SF. Whole-exome sequencing and whole-genome sequencing in critically ill neonates suspected to have single-gene disorders. Cold Spring Harb Perspect Med 6 2:a023168, 2015.
26. Hsu P, Xie Z, Frith K, Wong M, Kakakios A, Stone KD, Druey KM. Idiopathic systemic capillary leak syndrome in children. Pediatrics 135 3:e730–e735, 2015.
27. Xie Z, Chan ED, Long LM, Nelson C, Druey KM. High-dose intravenous immunoglobulin therapy for systemic capillary leak syndrome (Clarkson disease). Am J Med 128 1:91–95, 2015.
28. Dougherty GW, Loges NT, Klinkenbusch JA, Olbrich H, Pennekamp P, Menchen T, Raidt J, Wallmeier J, Werner C, Westermann C, et al. DNAH11 localization in the proximal region of respiratory cilia defines distinct outer dynein arm complexes. Am J Respir Cell Mol Biol 55 2:213–224, 2016.
29. Lucas JS, Adam EC, Goggin PM, Jackson CL, Powles-Glover N, Patel SH, Humphreys J, Fray MD, Falconnet E, Blouin JL, et al. Static respiratory cilia associated with mutations in Dnahc11/DNAH11: a mouse model of PCD. Hum Mutat 33 3:495–503, 2012.
30. Miyamoto Y, Yamauchi J, Sanbe A, Tanoue A. Dock6, a Dock-C subfamily guanine nucleotide exchanger, has the dual specificity for Rac1 and Cdc42 and regulates neurite outgrowth. Expl Cell Res 313 4:791–804, 2007.
31. Shaheen R, Faqeih E, Sunker A, Morsy H, Al-Sheddi T, Shamseldin HE, Adly N, Hashem M, Alkuraya FS. Recessive mutations in DOCK6, encoding the guanidine nucleotide exchange factor DOCK6, lead to abnormal actin cytoskeleton organization and Adams-Oliver syndrome. Am J Human Gen 89 2:328–333, 2011.
32. Kvarnung M, Nilsson D, Lindstrand A, Korenke GC, Chiang SC, Blennow E, Bergmann M, Stödberg T, Mäkitie O, Anderlid BM, et al. A novel intellectual disability syndrome caused by GPI anchor deficiency due to homozygous mutations in PIGT. J Med Gen 50 8:521–528, 2013.
33. Muramatsu F, Kidoya H, Naito H, Hayashi Y, Iba T, Takakura T. Plakoglobin maintains the integrity of vascular endothelial cell junctions and regulates VEGF-induced phosphorylation of VE-cadherin. J Biochem 162 1:55–62, 2017.
34. Xie Z, Chan EC, Yin Y, Ghosh CC, Wisch L, Nelson C, Young M, Parikh SM, Druey KM. Inflammatory markers of the systemic capillary leak syndrome (Clarkson disease). J Clin Cell Immunol 5:1000213, 2014.
35. Cabral RM, Liu L, Hogan C, Dopping-Hepenstal PJ, Winik BC, Asial RA, Dobson R, Mein CA, Baselaga PA, Mellerio JE, et al. Homozygous mutations in the 5’ region of the JUP gene result in cutaneous disease but normal heart development in children. J Invest Dermatol 130 6:1543–1550, 2010.

Capillary leak syndrome; critical illness; endothelial junctions; pediatrics; vascular permeability; whole-exome sequencing; CADD; combined annotated dependent depleted; EXaC; exome aggregation consortium; GAP; GTPase-accelerating protein; GEF; guanine nucleotide exchange factors; GnomAD; genome aggregate database; IRB; institutional review board; MAF; minor allele frequency; MGUS; monoclonal gammopathy of undetermined significance; NIAID; National Institute of Allergy and Infectious Disease; NIH; National Institute of Health; PCA; principal component analysis; PGDP; Pediatric Genomics Discovery Program; PPH2; PolyPhen-2; SCLS; systemic capillary leak syndrome; SIFT; Sorting Intolerant From Tolerant; SNV; single nucleotide variant; VEGF; vascular-endothelial growth factor; WES; whole-exome sequencing

Supplemental Digital Content

Back to Top | Article Outline
© 2019 by the Shock Society