Congenital chloride-losing diarrhoea and Crohn’s disease: a diagnostic and therapeutic challenge

Abstract

We describe the case of a patient with congenital chloride-losing diarrhoea (CCLD), global developmental delay and intermittent transaminitis who was diagnosed with Crohn’s disease after persistent anaemia and onset of rectal bleeding. CCLD is a rare autosomal recessive condition causing large-volume chloride-rich diarrhoea, metabolic alkalosis and potentially life-threatening electrolyte disturbance. A possible association between CCLD and inflammatory bowel disease (IBD) has recently become apparent; however, the underlying mechanism has not been identified, with the role of increased expression of tumour necrosis factor-alpha hypothesised. Early diagnosis and management are key for favourable outcomes within both CCLD and IBD, and understanding a potential link between the two conditions may lead to development of novel therapies and management strategies. We aim to highlight the pathophysiology, diagnosis and management of CCLD; its potential association with IBD; and the potential therapeutic difficulties within the management of patients with comorbid CCLD and IBD.

Case report

The patient was born in Saudi Arabia at 36 weeks to consanguineous parents. Maternal polyhydramnios and dilated bowel loops were noted on prenatal ultrasound scan. After a history of failure to thrive, chronic diarrhoea from birth unchanged by feeding adjustment and numerous admissions with dehydration and electrolyte imbalance, the patient was diagnosed with congenital chloride-losing diarrhoea (CCLD) aged 2 years and 9 months. Diagnostic faecal chloride was elevated at 144 mmol/L. At age 4, the family moved to Europe; follow-up was challenging due to their frequent relocation to Saudi Arabia. Genetic panel found a nonsense mutation (c.559G>T (p.(Gly187*))) of SLC26A3, previously reported in CCLD.¹ The patient’s CCLD was managed with potassium and sodium supplementation. His medical history also included global developmental delay, dysmorphic appearance, hypermobility and intermittent transaminitis. Laterally, the patient was noted to have a persistent iron deficiency anaemia requiring recurrent transfusion and intravenous iron.

He was seen in clinic aged 8 where routine bloods showed potassium 3.3 mmol/L, haemoglobin 47 g/L, white cell count 11×10⁹/L, platelets 522×10⁹/L, ferritin 2 µg/L, albumin 29 g/L, C-reactive protein 7 mg/L and erythrocyte sedimentation rate 110 mm/h. He was admitted to the gastroenterology ward for intravenous iron where his mother reported a week-long history of intermittent fresh red blood within his stool. Stool cultures were negative. Two samples for faecal calprotectin were raised at 219 µg/g and 448 µg/g.

The patient underwent upper gastrointestinal endoscopy and colonoscopy which revealed a small apthous ulcer at the duodenal cap and erythema and ulceration in the rectum, sigmoid, transverse and ascending colon (simple endoscopic score: 17) (figure 1). Pathology revealed mild active colitis within the sigmoid colon and moderate active inflammatory changes in the rectum consistent with the macroscopic diagnosis of Crohn’s disease. The patient underwent barium follow-through study as it was felt he would be unable to tolerate MRI imaging of his small bowel; this was unremarkable.

Figure 1.

Images from upper gastrointestinal endoscopy and colonoscopy demonstrating Crohn’s disease within the duodenum and sigmoid colon.

Oral prednisolone was commenced; it was felt that exclusive enteral nutrition (EEN) would not be feasible for induction of remission due to patient comorbidities. Mercaptopurine was commenced as maintenance therapy. Background diarrhoea (average six stools daily) persisted; however, rectal bleeding and nocturnal stooling stopped following treatment. Haemoglobin remained stable (95–105 g/L) for a period on oral iron supplementation. The patient went on to sustain a low-impact fracture of his foot with low bone mineral density noted on imaging. On last review, the patient remained clinically well on maintenance mercaptopurine and salt substitution.

Discussion

CCLD is a rare autosomal recessive condition caused by at least 55 mutations in the SLC26A3 (solute carrier family 26 member 3 or down-regulated in adenoma) gene^{1 2} on chromosome 7q31.³ Incidence is highest in Finland, Poland, Saudi Arabia and Kuwait^1–3 with more than 250 reported cases worldwide.^{1 3} The pathophysiology of CCLD involves defective chloride/bicarbonate and sodium/hydrogen exchange within the distal ileum and colon.^{2 3} This results in impaired absorption of fluid, chloride (hypochloraemia)² and sodium (hyponatraemia);³ impaired excretion of bicarbonate (metabolic alkalosis);² activation of the renin–angiotensin system^{2 3}; and increased sodium reabsorption in the distal colon and consequent hypokalaemia.³ The condition causes large-volume chloride-rich diarrhoea of prenatal onset, resulting in polyhydramnios, dilated intestinal loops on prenatal ultrasound and prematurity;^{2 3} all occurring within the aforementioned case. Postnatally, CCLD may present like obstruction with abdominal distension and lack of meconium.³ The diarrhoea associated with CCLD may resemble urine and therefore may be difficult to identify.^{2 3} Growth and development may be delayed as a result of recurrent fluid and electrolyte depletion. Causing potentially life-threatening electrolyte imbalance and dehydration, CCLD is considered a medical emergency; it is usually lethal within the first months of life if left untreated.^{2 3}

A range of extraintestinal manifestations may also be present in CCLD including chronic kidney disease, male subfertility and elevated sweat chloride.^{2 3} Inguinal hernias have been found to be 10 times more common in children with CCLD, thought to be due to raised intra-abdominal pressure.²

Diagnosis is made based on raised faecal chloride (>90 mmol/L).² This must be assessed when fluid and electrolytes are balanced as if deplete, diarrhoea may decrease and produce a false-negative result. Genetic testing may be undertaken to identify the specific mutation in SLC26A3.³

Early diagnosis and management are crucial for favourable outcomes and are associated with normal growth and development. The case we describe was diagnosed relatively late in life (2 years 9 months) after recurrent episodes of dehydration and electrolyte imbalance; this may have contributed to developmental delay. Lifelong salt and volume substitution—as prescribed to the patient described above—are the mainstay of treatment.² Patients are vulnerable to deterioration during intercurrent illness—this must be promptly and aggressively managed with extra fluid and electrolyte replacement, avoiding bicarbonate-containing fluids for rehydration.³ Electrolytes and renal function should be regularly monitored alongside urine chloride. If salt repletion is inadequate, diarrhoea will decrease and chloride will be reabsorbed in the distal colon/nephron and consequently urine chloride excretion is poor.³ Therapies to increase chloride absorption are generally ineffective in the long term.²

CCLD and inflammatory bowel disease (IBD)

The potential association between CCLD and IBD has been previously reported; however, the underlying mechanism remains unidentified. A 2016 report following six patients with CCLD identified two patients with comorbid IBD and found no difference in gestational age, birth weight or stool chloride level at diagnosis between those with IBD and the others.⁴ Retrospective analyses of 36 Finnish patients with CCLD described one patient with Crohn’s disease and another with unspecified colitis.²

SLC26A3 has been found to be downregulated in inflamed colonic mucosa in ulcerative colitis.³ It has recently been hypothesised that the gene may act reciprocally with tumour necrosis factor-alpha (TNF-α) to mediate gut inflammation; in one study, TNF-α appears to reduce SLC26A3 expression and SLC26A3 downregulation appears to increase mRNA and protein expression of TNF-α in cell culture medium.⁵

IBD associated with CCLD presents a diagnostic and therapeutic challenge. Presence of diarrhoea at baseline complicates the initial assessment of IBD, and its persistence and frequency through IBD treatment eliminates this as a reliable marker of IBD activity. Other indicators (such as nocturnal stooling, blood within stool or abdominal pain as within the case described above) may be used in such patients. In this case, the patient’s developmental delay may have led to later presentation of Crohn’s disease, further complicated disease assessment and also influenced the decision to use steroids rather than EEN for induction of remission. Given the family background of this patient, other genetic causes of IBD and indeed other genetic disorders may be present; further genetic screening has not, however, been undertaken currently due to family non-attendance at a scheduled genetics clinic appointment.

It is possible that IBD in patients with CCLD is underdiagnosed due to this diagnostic difficulty. A pan-European study is now being undertaken to understand the relationship between CCLD and IBD in more detail (Norsa et al,⁴ personal communication), and may lead to development of novel therapies and diagnostic and management strategies for both conditions.