Chronic Megacolon Presenting In Adolescents or Adults: Clinical Manifestations, Diagnosis and Genetic Associations

Abstract

Chronic megacolon is rarely encountered in clinical practice beyond infancy or early childhood. Most cases are sporadic, some are familial, and present from adolescence to adulthood. There is a need for diagnostic criteria and opportunity to understand the potential role of genetic variants in the development of these cases of non-Hirschsprung’s megacolon. This article reviews the clinical manifestations, current diagnostic criteria, and explores the use of intraluminal measurements of colonic compliance to confirm the diagnosis when the radiological imaging is not conclusive. The diverse genetic associations with chronic acquired megacolon beyond childhood are reviewed including recent observations of the potential association of SEMA3F gene in a family with megacolon.

Introduction

The genetic association of congenital megacolon (Hirschsprung disease) due to aganglionosis of the colon is well described; however, the potential genetic basis of familial chronic megacolon presenting in adolescence or adulthood remains unclear. Due to the rarity of the condition, it remains unclear whether this is a single condition or a heterogeneous group of diseases with a common phenotypic end point. However, the clinical phenotype of this, albeit rare, disease is fairly clear from the descriptions in the literature: this is a distinct form of megacolon which presents beyond adolescence and is not the result of organic disease such as in the case of toxic megacolon in association with fulminant inflammatory bowel disease, or megacolon due to distal stenosis, hypoganglionosis or aganglionosis of the rectum, or megacolon secondary to infection.

Clinical Features of Chronic Acquired Megacolon

Symptoms of acquired megacolon differ from those of congenital megacolon. Adults with megacolon generally present with constipation, abdominal pain, distension and discomfort from abdominal gas. While there is likely an overlap with those diagnosed with constipation-predominant irritable bowel syndrome (IBS-C) [1], it is unclear what percentage of patients presenting with IBS-C actually have megacolon. On the other hand, children with megacolon generally present with meconium ileus in the neonatal period (usually in association with Hirschsprung disease), with colonic obstruction during infancy, or with fecal impaction and soiling during childhood. The latter is typically the result of functional fecal retention [2].

The presence of megacolon can be complicated by the development of obstruction, fecal impaction, or volvulus [3]. There is an association of acquired megacolon with neuropsychiatric conditions including schizophrenia and mental retardation, as well as organic central nervous system conditions like cerebrovascular disease and epilepsy [3]. Rates of complications are also higher among those with a concomitant psychiatric condition [4].

Investigations may show high resting anal canal pressures, high incidence of very slow anal pressure waves, and decreased rectal sensation with increased rectal capacity that may reflect either megarectum or, sometimes, independent co-morbidity in the form of rectal evacuation disorders which have been observed in sporadic and in MEN2B-associated megacolon [5]. Colonic transit times are generally prolonged (Figure 1). Colonic intraluminal recordings document increased colonic compliance (Figure 2) and reduced colonic tone, but a retained phasic contractile response to feeding [5].

Figure 1.

Example of dilated colon due to chronic megacolon on coronal CT imaging and retardation of colonic transit in a 43 year old woman

Figure 2.

Colonic compliance in (A) healthy controls, (B) functional constipation/constipation-predominant irritable bowel syndrome and (C) diarrhea-predominant irritable bowel syndrome (IBS-D) control groups; and (D) patients with chronic megacolon. Note the markedly increased volume of the intracolonic balloon (10cm long) in patients with megacolon compared to controls. Note also the marked increase in intraballoon volume (>300mL) at 16mmHg distension in all except one patient with megacolon, which is observed in only one healthy control and in none of the IBS-D patients. Reproduced from ref. 5, O’Dwyer, et al. Dig Dis Sci 2015;60:2398–2407.

Acquired megacolon remains a poorly understood disease. Our aims were to assess clinical features, diagnostic criteria, pathophysiology, and genetic associations in patients presenting with megacolon in adolescence or adulthood.

Current Diagnostic Criteria for Chronic Acquired Megacolon

In the past, there have been diverse criteria proposed for the diagnosis of megacolon. Studies reported in 1985 utilizing double contrast barium enema established a normal rectosigmoid diameter at the level of the pelvic brim as <6.5cm. The cut-off for a sigmoid diameter >10cm on radiology is based on three studies which calculated the average sigmoid diameters of patients with acquired megacolon as 10cm with standard deviation ranging from 2 to 3.5cm in different studies. Other studies of acquired megacolon have utilized the enema-filled sigmoid colon rising above the iliac crest or symptom recurrence after segmental colectomy as diagnostic criteria. Importantly, these measurements were taken at a specific level of the sigmoid and utilized enema contrast studies. Studies have shown that single plain radiographs, which are often used for diagnosis of megacolon, have significant variability and that the location of the measurement is often imprecise. In addition, utilizing solely sigmoid measurements may conceivably miss a subset of patients who present with isolated proximal colon dilatation [6]. Coronal images on CT scans of the abdomen and pelvis may provide a more thorough assessment of diameter of the entire colon (Figure 2).

A recent systematic review of acquired megacolon has attempted to establish diagnostic criteria for chronic idiopathic megacolon presenting in adolescence or adulthood: 1.exclusion of organic disease, 2. sigmoid diameter ≥10cm on radiological studies, and 3. presentation with symptoms of constipation, distension, abdominal pain, and/or gas distress.

Exclusion of organic disease involves exclusion of infectious etiology, documentation of an intact anorectal inhibitory reflex (RAIR), and absence of hypoganglionosis on rectal biopsy. There may be impaired RAIR which is often due to the large diameter of the rectum (due to associated megarectum), and the histological findings showing retained cholinergic neurons is more relevant than the absent RAIR. In contrast with the preservation of ganglion cells in the submucosal region on rectal biopsy, there may be significant histological abnormalities in the deeper layers (typically on the resected specimen) such as decreased interstitial cells of Cajal (ICC) [7,8], diminished myenteric ganglia [9], diminished enteric neural densities [10], and/or enteric smooth muscle hypertrophy [11]. These histological changes may precede development of clinical symptoms, as similar histopathology can be seen in dilated and non-dilated colonic segments [9].

Intracolonic Compliance Measurements to Identify Chronic Acquired Megacolon

Given the heterogeneity and limitations of current diagnostic measures, we evaluated the use of colon compliance on intracolonic measurements with a barostatically-controlled polyethylene balloon placed transanally into the left colon. Figure 2 and Table 1 show data from 117 healthy volunteers and 10 patients with megacolon evaluated with the intracolonic balloon, as well as the estimated median diameter of the left colon, assuming the balloon tied at both ends to a tube conforms to the shape of a cylinder [5]. This may be a more objective test for measuring increased colonic compliance in megacolon, and it could be standardized, simplified by measuring volume at 20, 32 and 44mmHg distension, and may be generalizable for clinical and research use.

Table 1.

Colonic compliance in 117 healthy controls and 10 patients with megacolon, based on volumes (mL) of 10cm-long balloon in the left colon at different balloon pressures

Pressure, mmHg	Median volume, mL	10%ile volume, mL	90%ile volume, mL	Estimated median diameter, cm	Median volume, mL	10%ile volume, mL	90%ile volume, mL	Estimated median diameter, cm
HEALTHY VOLUNTEERS					PATIENTS WITH MEGACOLON
20	116.0	81.0	195.0	3.84	357.0	273.5	442.5	6.74
32	194.0	131.0	278.8	4.97	465.0	355.0	520.0	7.70
44	246.0	180.7	332.4	5.60	566.0	397.6	600.7	8.49

Familial Clustering and Genetic Associations of Acquired Megacolon

Clinical series have documented familial clustering of acquired megacolon, suggesting that altered genetics may be implicated in the development of the condition. Gene mutations implicated in the development of congenital megacolon and Hirschsprung disease include variants or mutations in RET kinase, endothelin receptor B (EDNRB), NRG1 receptor ERBB2, and RET co-receptor GFRA1 which are involved in enteric nervous system development and are known as the “damaging rare variants” [12]. However, several variants in RET that were predicted to be benign were also found in higher numbers in those with congenital megacolon, suggesting additive effects of “mild” variations which reduce RET expression and lead to disease in those with susceptible backgrounds [13].

RET mutations have also been associated with megacolon presenting in adolescence or adulthood in multiple endocrine neoplasia syndrome type 2B (MEN 2B). However, such a phenotype (with 100% medullary carcinoma of the thyroid, a typical facies, with blubbery lips and ganglioneuromas on the tongue) is rarely available in the majority of adults with acquired megacolon.

Despite several genetic mechanisms identified in the development of Hirschsprung disease and the involvement of RET mutations in megacolon development as part of the MEN 2B, a single genetic mutation has not been conclusively demonstrated for the pathogenesis of acquired megacolon. Here we review other candidate genes reported in the literature to date:

1. ACTG2 gene codes for gamma-2 actin, a member of the actin protein family. Mutations of ACTG2 have been associated with megacystis in chronic intestinal pseudo-obstruction. Patients reported in case series, including a large Swedish family with mutation of ACTG2, presented with symptoms termed “pseudo-Hirschsprung” disease, but with phenotypes indicating megacystis, microcolon, and intestinal hypoperistalsis, and not megacolon [14,15].

2. GFRA1 codes for the glial cell line-derived neurotrophic factor (GDNF) Family receptor alpha-1, which binds GDNF and mediates the activation of the RET protein tyrosine kinase. GDNF knock-out mice present with lack of enteric neurons throughout the digestive tract [16]. GFRA1 was also shown to be among the genes upregulated in rat gut neural crest stem cells [17], and under expression of this gene causes Hirschsprung disease and enterocolitis in mice [18]. To date, there have not been any clinical reports of GFRA1 alterations causing megacolon with onset in adults or adolescents.

3. NKX2–1 codes a transcription factor that is involved in early development of several organs including thyroid, lung, and forebrain. NKX2–1 also interacts with the RET promoter to impact RET transcription. Polymorphisms in the RET promoter region overlapping with the NKX2–1 homeobox binding motif disrupts the interaction of NKX2–1 with the RET promoter and have been associated with increased risk of Hirschsprung disease [19]. Thus, alterations of NKX2–1 could lead to the development of megacolon. A single study describing NKX2–1 as a Hirschsprung disease locus was subsequently retracted [20].

4. KIF26A (Kinesin Family member 26A) is involved in ATP hydrolysis allowing microtubule stability [21]. Over expression of this gene seems to suppress GDNF-RET signaling by binding to a protein (GRB2) required for assembly of RET signaling complexes. Mice homozygous for KIF26A knockout died within 5 weeks of birth with emaciation, growth retardation, megacolon, and enteric nerve hyperplasia causing disturbed bowel functions [21]. No reports of KIF26A mutation has been described in humans with megacolon to date.

5. The tropomyosins are a highly conserved family of actin binding proteins that play important roles in a wide range of cellular processes. In humans, alternate splicing of four tropomyosin genes creates three skeletal muscle isoforms; mutations in three tropomyosin genes have been associated with four different disorders of striated muscle and cardiomyopathy: TPM3 encodes for alpha-tropomyosin 3, which is mostly expressed in slow, type 1 muscle fibers (as opposed to type 2 fast-twitch muscle fibers. Mutations of this gene have been associated with congenital muscular dystrophy [22] and atypical nemaline myopathy 1 [23]. TMP3 was studied as a muscle marker in patients with idiopathic megarectum and megacolon using anti-rabbit or anti-mouse antibodies. This study did not find any abnormal staining of tropomyosin, but did note reduction in beta actin and myosin light chain kinase in the muscularis externa [24]; there was no documentation of TPM3 genetic mutation in a patient with megarectum and megacolon.

SEMA3F Gene Variant (rs774797321) in Familial Megacolon

Our group recently evaluated multiple gene candidates in a family with five generations and 49 members, six of whom had chronic megacolon presenting in adolescence or adulthood (Figure 2), 1 with megaduodenum, 7 who experienced constipation, and 16 without GI symptoms. Exome DNA sequencing of two patients with overt megacolon evaluated the genetic variants with functions related to neuroblast migration, differentiation, maturation or survival (including all those described in Hirschsprung disease); no definite association was found. In 30 family members who provided DNA samples, we then performed Sanger DNA sequencing of specific genes of interest described above that had possible associations with megacolon including GFRA1, NKX2–1, KIF26A, TPM3, and ACTG2; none of these showed definite associations with the patients with megacolon or other family members. However, exome DNA sequencing in all 6 family members with megacolon and 1 member without megacolon revealed 12 genetic variants in 11 genes which co-segregated completely with patients with megacolon. These genes were evaluated in the literature for associations with enteric nervous system development [25].

Among these genes, two genes had potential in development of megacolon. Rad21 codes for a protein involved in regulating, structuring, and organizing chromosomes during cell division. A single nucleotide variant of Rad21 has been previously reported to be associated with Mungan syndrome, characterized by familial visceral neuromyopathy, pseudo-obstruction, megaduodenum, Barrett esophagus, and cardiac abnormalities in a large consanguineous Turkish family [26]. In the family evaluated at Mayo Clinic, a different Rad21 variant was identified, and it was not consistent with the pathological alteration reported in the Turkish family. The second gene, SEMA3F, is involved in axon guidance in neural development and is expressed during development of the enteric nervous system in embryonic mice. More importantly, several other members of the SEMA family (SEMA3A, SEMA3C, and SEMA3D at SNPs rs1583147, rs12707682, and rs11766001) [27–29] have been associated with non-syndrome Hirschsprung disease.

Bioinformatics analysis of protein-protein interaction data identified other genes encoding proteins associated with the functions of the SEMA3F gene. The two most relevant are KIT (proto-oncogene receptor tyrosine kinase) and TEK receptor tyrosine kinase which are associated with neural crest migration to primordial gut or in development of precursors of two gastrointestinal pacemakers that regulate motility patterns [30,31], that is, interstitial cells of Cajal [c-Kit(+)] and PDGFRA(+) fibroblast-like cells. From the genetic analysis of this pedigree of familial megacolon, further evaluation into the role and presence of SEMA3F variants should be explored as a contributor to development of acquired chronic megacolon.

Conclusions

Despite increasing understanding of the pathophysiology and diagnosis of congenital megacolon, the mechanisms leading to adolescent and adult onset chronic megacolon remain unclear. Recent reviews have emphasized enlarged sigmoid diameter (≥10cm) and characteristic symptoms for diagnosis of chronic megacolon in the absence of organic disease such as obstruction or acute colitis. However, radiological diagnosis is limited by potential day to day variations, technical variability of plain radiographs, and unstandardized location of measurement or missing isolated proximal megacolon. Thus, coronal images on CT scans of the abdomen and pelvis may provide a more thorough assessment of diameter of the entire colon (Figure 2). When the diagnosis is not overt on radiological imaging, an intracolonic measurement of colonic compliance in the affected segment provides an objective measurement of megacolon, as summarized in Table 1. In addition, while there seems to be a familial phenotype in the development of chronic megacolon, little is known about the possible genetic mechanisms and whether this is a separate phenotype from acquired, sporadic (non-familial) chronic megacolon. We also propose further investigation into SEMA3F as a possible causative gene, based on results of genetic analysis in a pedigree of familial megacolon.

Figure 3.

Examples of abdominal plain radiographs or CT images in coronal section to illustrate the large colonic diameters in 6 members of a family with megacolon. Reproduced from ref. 25, Camilleri M, Wieben E, Eckert D, Carlson P, Hurley O’Dwyer R, Gibbons D, Acosta A, Klee EW. Familial chronic megacolon presenting in childhood or adulthood: Seeking the presumed gene association. Neurogastroenterol Motil. 2019 Jan 20:e13550. doi: 10.1111/nmo.13550. [Epub ahead of print]