Congenital sodium diarrhoea caused by rare de novo activating guanylate cyclase mutation
- 1 Neonatology, Advocate Children's Hospital, Park Ridge, Illinois, USA
- 2 Pediatric Gastroenterology, Advocate Children's Hospital, Park Ridge, Illinois, USA
- 3 Pediatric Gastroenterology, NorthShore University HealthSystem, Evanston, Illinois, USA
- Correspondence to Dr Surichhya Bajracharya; suribajracharya@gmail.com
Abstract
A male infant with prenatal history significant for polyhydramnios requiring multiple amnioreductions with suspicion of small bowel atresia was born at 31 weeks 5 days’ gestation with abdominal distension. He underwent three exploratory laparotomies and ileostomy for small bowel obstruction and was found to have fluid-filled intestinal dilatation. Serum and stool chemistries suggested sodium secretory diarrhoea. A rapid whole-exome sequencing confirmed de novo guanylate cyclase mutation variant as a cause for his congenital sodium secretory diarrhoea. He required large volume of fluid and electrolyte replacement along with total parenteral nutrition. Several medications to restore normal sodium homeostasis by targeting molecular mechanisms and pathogenesis described in previous literature failed to decrease stool output and electrolyte loss. He was discharged home at 11 months of age on total parenteral nutrition and weekly iron therapy.
Background
Congenital sodium diarrhoea (CSD) is a rare type of secretory diarrhoea caused by impaired intestinal sodium absorption leading to increased sodium and water flux towards intestinal lumen. Three genes—SPINT2, SLC9A3 and GUCY2C—have been identified in the molecular aetiology of CSD. A syndromic CSD caused by loss-of-function mutations in SPINT2 is associated with extraintestinal manifestations such as choanal stenosis, hypertelorism, corneal erosions, cleft palate and polydactyly. A non-syndromic CSD with no extraintestinal manifestation has been reported and caused by either SLC9A3 mutations, resulting in absent or non-functional apical sodium hydrogen antiporter 3 (NHE3), or by constitutively hyperstimulating mutations in receptor guanylate cyclase C (GC-C) encoded by the GUCY2C gene. This results in downregulation of NHE3 by elevated intracellular cyclic guanosine monophosphate (cGMP) levels. NHE3 absorbs a major part of intestinal sodium and is important for systemic volume and acid-based homeostasis.1
Less than 50 cases of CSD have been reported after the first case report in 1985.2–5 We report a novel variant of heterozygous de novo mutation in GUCY2C causing CSD. To our knowledge, this variant has not been reported in the literature or in a large population database. Published reports on CSD have described the molecular mechanism and pathogenesis of this disease process with none describing medications to restore normal sodium homeostasis by targeting the molecular mechanism and pathogenesis described previously.
Case presentation
A male infant was born at 31 weeks 5 days’ gestational age to a G2P1 Asian mother via caesarean section for non-reassuring fetal heart rate. Prenatal course was significant for polyhydramnios that required five amnioreductions with high suspicion of fetal small bowel obstruction. Aneuploidy testing from amniotic fluid, fetal echocardiogram and prenatal labs were unremarkable. The mother received steroids for fetal lung maturity. His APGAR scores were 7 and 9 at 1 and 5 min of life, respectively. Birth anthropometry was appropriate for age with no dysmorphism except for severely protuberant abdomen (figure 1). An abdominal X-ray revealed dilated intestine (figure 2) and an ultrasound of the abdomen demonstrated numerous dilated fluid-filled intestine (figure 3). A lower gastrointestinal tract contrast radiography demonstrated a non-obstructed slightly smaller calibre colon. He was kept nil per oral with total parenteral nutrition (TPN). His abdominal distension progressed with metabolic acidosis (HCO3 8 mmol/L, normal reference value 19–24 mmol/L)6 and hyponatraemia (sodium (Na) 122 mmol/L, normal reference value 130–145 mmol/L).6 Exploratory laparotomy on day of life (DOL) 5 for possible small bowel obstruction discovered an adhesion traversing the cecum, the fourth part of the duodenum and proximal jejunum as well as a kinked piece of terminal ileum, which appeared to be the distal source of obstruction. His abdominal distension recurred with dilated intestinal loops on abdominal imaging with continued metabolic acidosis and hyponatraemia. A second exploratory laparotomy performed on DOL 8 revealed fluid-filled dilated bowel from ligament of Treitz down to the rectum. An end ileostomy and creation of ileal mucous fistula were performed along with small and large intestinal biopsy with the provisional diagnosis of partial functional small bowel obstruction. He subsequently underwent a third exploratory laparotomy for bowel obstruction at 8 weeks of life and was found to have an internal hernia with volvulised segment of necrotic ileum measuring approximately 8 inches. At that time, he underwent lysis of adhesions, bowel resection and revision of ileostomy and mucous fistula.
Image demonstrating distended abdomen due to fluid retention.
Abdominal X-ray demonstrating dilated intestinal loops.
Abdominal ultrasound demonstrating dilated fluid-filled intestinal loops.
Investigations and differential diagnosis
Analysis of output from ileostomy demonstrated secretary diarrhoea with large volume of watery stool with osmotic gap of <50 mOsm/kg (stool osmotic gap <50 mOsm/kg indicates secretary diarrhoea and stool osmotic gap ≥100 mOsm/kg indicates osmotic diarrhoea)7 even during fasting and normalisation of serum Na with electrolyte replacement. The high stool sodium (130–140 mmol/L, normal reference value 30 mmol/L), stool chloride (>110 mmol/L, normal reference value 10–20 mmol/L)8 and low stool potassium (3 mmol/L, normal reference value 75 mmol/L) along with alkaline stool pH of 8 and metabolic acidosis confirmed that the secretary diarrhoea was due to CSD rather than chloride diarrhoea. Intestinal histopathology and immunohistochemistry were normal excluding microvillus inclusion disease, tufting enteropathy, visceral myopathy, congenital absence of interstitial cells of Cajal and Hirschsprung disease. Rapid whole-exome sequencing sent on DOL 11 reported a heterozygous mutation in the GUCY2C gene as a de novo variant designated c.2324T>G, predicted to result in the amino acid substitution p.Leu775Arg causing autosomal dominant CSD. The variant was not found on genetic testing of both parents. Parents were counselled that it is unlikely that they will have additional affected children and our patient has 50% risk of passing on the variant to each child our patient may have in the future.
Treatment
Daily stool volume and weekly serum and stool electrolyte evaluations were done throughout the hospital course (figure 4). He had an average stool output of 120 mL/kg/day (stool output of >10 g/kg/day is considered as diarrhoea)7 with Na requirement of 21 mEq/kg/day from both TPN and replacement fluid. He developed TPN-induced cholestasis, which resolved after treatment with ursodeoxycholic acid, phenobarbital and cycled fish oil-based lipid (Omegaven). Minimal enteral feedings of breast milk/amino acid-based formula were initiated at fourth week of life. Enteral feeding could not be advanced beyond 20 mL/kg/day as stool output subsequently increased creating electrolyte instability. Pureed foods were started at 5 months of age; however, he showed poor interest and motivation for oral feeding. He had appropriate weight gain and was receiving about 72 kcal/kg/day through TPN at the time of discharge.
Serum and stool electrolytes along with ileostomy output during treatment with different medications (this figure was created by one of the authors, Surichhya Bajracharya).
After extensive discussion with the parents, neonatology, pharmacology, gastroenterology and international experts in this condition, several medications were tried (table 1). There were no major adverse effects associated with these medications. He was discharged home at 11 months of age on home TPN and weekly iron therapy with possible need for intestinal transplant in future.
List of medications the patient received
| Name of the medication | Mode of delivery | Interval of initiation from birth | Dose | Duration of treatment | Decrease in stool output or electrolyte loss/comments |
| Octreotide | Intravenous | 13 days | 2 µg/kg/hour Increased to 10 µg/kg/hour |
4 months | No |
| Clotrimazole | Nasogastric | 6 weeks | 10 mg every 8 hours Increased to 25 mg every 6 hours |
8 weeks | No Held during laparotomy and was restarted at 11 weeks at dose of 20 mg every 6 hours Transient elevation of liver enzymes |
| Methylprednisolone | Intravenous | 11 weeks and 14 weeks | 1 mg/kg/day | 3 days | No |
| Methylene blue | Intravenous | 5 months | 1 mg/kg/day | 2 days | No Monitored for haemolytic anaemia. Haemolysis was not seen on complete blood cell count with reticulocyte count |
| Butyric acid | Nasogastric | 8 months | 25 mg/kg/day Increased to 100 mg/kg/day |
3 weeks | No |
| Clonidine | Transdermal | 9 months | 0.5 mg/week Increased to 1 mg/week |
2 weeks | No Transient hypotension, hypothermia and decrease in activity initially which resolved during subsequent days of treatment |
| Loperamide | Nasogastric | 10 months | 0.75 mg/kg/day Increased to 1.5 mg/kg/day |
10 days | Minimal decrease in stool output from 120 mL/kg/day to 90 mL/kg/day Medication stopped for fear of fluid retention with no decrease in stool electrolyte loss |
Outcome and follow-up
On follow-up at 2 months from discharge, his developmental milestones and growth percentile were appropriate. He was receiving 15 mL of thickened formula two times per day from recently placed G-tube and continued to have oral aversion with minimal oral intake of about 100 mL per day of whole milk and baby food. His stool output was stable at 130 mL/kg/day and he was on TPN with normal serum electrolytes.
Discussion
CSD is a clinically and genetically heterogeneous rare disease. Patients typically present with a history of polyhydramnios, premature birth between 32 and 35 weeks’ gestational age, appropriate weight and length, and abdominal distension. Watery diarrhoea is not observed initially either due to severe dehydration or intestinal paralysis from intestinal smooth muscle relaxation due to cGMP leading to pseudo-obstruction with dilated fluid-filled loops of intestine, which may result in volvulus. In our patient, polyhydramnios requiring multiple amnioreductions and dilated fluid-filled intestinal loops could possibly have been due to intestinal paresis leading to functional intestinal obstruction. Watery diarrhoea was not obvious until the ileostomy was created. In CSD, the affected neonate typically undergoes explorative laparotomies.1 9 In one report, 8 out of 32 Norwegian family members with familial diarrhoea syndrome caused by an activating GUCY2C mutation underwent laparotomy from small bowel obstruction and some family members required a second or even third laparotomy because of recurrent obstruction.9 All of the patients with CSD required TPN for treatment of dehydration, Na supplementation and maintenance of normal body growth for at least several months.1 Corneal erosions and peripheral retinal drusen-like deposits have been described in syndromic CSD.10 The normal morphology and ophthalmological examination along with the type of mutation are suggestive of non-syndromic CSD in our patient. In the Norwegian family, diarrhoea was described as mild and started in infancy.9 More severe phenotype with diarrhoea beginning prenatally was described in four patients with dominant gain-of-function GUCY2C mutation leading to non-syndromic CSD.11 All four patients harboured different heterozygous de novo mutations in GUCY2C with one of the mutations being c.2324T>C, p.Leu775Pro. Our patient has a novel variant of mutation with different nucleotide substitution at this position designated c.2324T>G, p.Leu775Arg.
Our patient was treated with several medications in agreement with the parents in an attempt to restore normal Na homeostasis by targeting proposed molecular mechanisms and genetic mutation described in previous literature.
Effective use of octreotide has been described in several case reports for intractable diarrhoea.12 13 Octreotide mimics natural somatostatin and decreases secretion of gastrin, vasoactive intestinal peptide, insulin, glucagon, secretin, motilin, pancreatic polypeptide and splanchnic blood flow.
Antifungal drug clotrimazole has been demonstrated to be a potent blocker of basolateral cAMP and Ca2+ gated K+ channels in enterocytes and therefore potentially has therapeutic efficacy for secretory diarrhoeas.14
Glucocorticoids have been found to stimulate NHE3 receptor by genomic regulation of NHE3 resulting in increased NHE3 mRNA level in animal study.15 A link of CSD with dominant GC-C mutation with inflammatory bowel disease also indicated possible effectiveness of steroid usage.1 9 11
Methylene blue, which has not been used for diarrhoea in the past, was used due to its inhibitor effect on nitric oxide synthase and guanylate cyclase.16 17
We used increasing doses of butyrate from 25 to 100 mg/kg/day for 3 weeks based on a case report on a patient with congenital chloride diarrhoea who showed improvement with decrease in stool frequency, volume and stool electrolyte loss following treatment. The proposed mechanism of butyrate is stimulation of an electroneutral Na chloride absorption activated by parallel chloride/butyrate and NHE3 with secondary upregulation of the NHE3 and chloride/HCO3 exchangers.18
Clonidine as an adrenergic agonist inhibits cyclic monophosphate that could inhibit intestinal secretion. Transdermal administration of clonidine 0.3 mg per week was associated with reports of decrease in stool output and Na loss after 7 days in patients with short bowel syndrome and high-output proximal jejunostomy.19 20
In another case report by Kidowaki et al, loperamide decreased stool output with an increase in serum Na when used in CSD with no genetic diagnosis. The author stopped loperamide treatment for acholic stool of unknown mechanism and observed increase in stool output after stopping of loperamide.3 Loperamide binds to the opiate receptor in the gut wall inhibiting the release of acetylcholine and prostaglandins reducing propulsive peristalsis, daily stool volume and loss of fluid and electrolytes.21 Loperamide was not used earlier in our patient because inhibiting acetylcholine could worsen intestinal motility and exacerbate abdominal distension. Philip in 1977 described occupancy of muscarinic acetylcholine receptors stimulates a guanylate cyclase in neuroblastoma cells.22 We used loperamide later at infancy when all the medications failed. A slight decrease in stool output from 120 mL/kg/day to 90 mL/kg/day was seen. However, the infant appeared oedematous with rapid weight gain. Loperamide was stopped out of concern for fluid retention with no decrease in stool electrolyte loss.
All of the above-mentioned medications were deemed ineffective given no consistent decrease in stool output or electrolyte loss (figure 4). Ileostomy closure and reconnection was not performed in our patient before discharge due to presence of a very high ileostomy output which could jeopardise the quality of life with persistent large-volume diarrhoea. Also, refeeding from distal mucous fistula was not done due to uncertainty regarding colonic function as fluid-filled dilated loops of large intestine were seen during laparotomies.
In the Norwegian family, among those who had diarrhoea that started in infancy, diarrhoea fairly remained constant over the years and in some subsided by middle age.6 Among four cases reported by Muller et al, all cases required parenteral fluids for at least 2 years and two required ongoing TPN; three of them had normal outcome and remaining one with mutation at similar position c2324T>C as our patient required partial small bowel resection with colitis.11 Some cases required chronic nutritional support or small bowel transplantation. Our patient required TPN for nutrition, fluid and electrolyte replacement and had appropriate developmental milestone for corrected gestational age at the time of discharge.
In recent years, compounds that block cGMP-dependent protein kinase II23 have been described on intestinal organoids and phenylpyrimidines as GC-C inhibitors on human colorectal carcinoma cells.24 The treating team was not aware of these studies during inpatient treatment of our patient. These compounds could be possibly tried in our patient or any similar case of CSD in future after acquiring more information on these compounds.
Conclusion
This case report describes our approach to the diagnosis and management of a rare novel mutation producing CSD. In neonates with unexplained chronic diarrhoea, whole-exome sequencing should be considered as part of the initial work-up. All possible treatments targeting proposed mechanism disrupted by the mutation were trialled but failed to decrease stool output or electrolyte loss. Further studies on a molecular level with potential therapy to maintain electrolyte homeostasis are needed for this rare genetic disease. We conclude that the patient should have good prognosis if electrolyte homeostasis can be achieved.
Learning points
-
Congenital sodium diarrhoea is a rare disease.
-
This is a case with a non-syndromic congenital sodium diarrhoea caused by a rare variant of de novo activating guanylate cyclase mutation, which has not been reported in the literature or in a large population database.
-
High clinical suspicion based on clinical presentation, serum and stool chemistry along with genetic evaluation is necessary for early diagnosis and appropriate fluid, electrolyte and nutrition management to prevent severe dehydration, hyponatraemia, metabolic acidosis and malnutrition.
-
Various treatments targeting the proposed molecular mechanisms and genetic mutations described in previous literature failed to restore normal fluid and electrolyte homeostasis in this case, highlighting the need for more research targeting the disease mechanism and potential treatment.
Ethics statements
Patient consent for publication
Acknowledgments
The authors would like to acknowledge Dr Prazad for her critical review of the case report and the entire NICU team, paediatric gastroenterology and genetics team at Advocate Children’s Hospital, Park Ridge, who have helped with the management of the case.
Footnotes
-
Contributors SB—active participation in the treatment of the case with substantial contribution to the literature search, conception or design of the work, acquisition of data, drafting the work and revising it critically for important intellectual content. DS—active participation in the treatment of the case with substantial contribution to drafting the work and revising it critically for important intellectual content. JB—active participation in the treatment of the case with substantial contribution to the literature search, drafting the work and revising it critically for important intellectual content. VB—active participation in the treatment of the case with substantial contribution to the literature search, drafting the work and revising it critically for important intellectual content.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
-
Competing interests None declared.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2022. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer