Background

For a condition that is sometimes treated surgically, Acquired Megacolon (AMC) is poorly understood and diagnostic criteria remain obscure [1,2,3,4,5,6]. The term refers to a colon of increased diameter and increased length, in the absence of organic disease [7,8,9]. The colon is often described as distended or dilated [10,11,12,13]. Some authors differentiate ‘colonic redundancy’ from ‘megacolon’, by describing increased colonic length with reduplication, opposed to a colon of increased diameter [9, 14,15,16]. However, increased length and diameter often co-exist. The definition AMC used in this paper encompasses both increased colonic length and diameter with negligible rectal involvement where possible.

Many symptoms including constipation, distension, abdominal pain and a poor sense of wellbeing are attributed to AMC [17]. Pathophysiology, natural history and effective symptom management, although speculated, are unknown [3, 7, 16, 18,19,20,21,22,23,24]. A large body of evidence exists, including literature reviews, which have analysed surgical outcomes for AMC and the diagnosis of idiopathic megabowel (megarectum-inclusive disease) [1, 3, 4]. However, consensus on diagnostic criteria is still lacking.

The objective of this systematic literature review is to refine diagnostic criteria for this condition and to evaluate symptoms and pathophysiology that may be associated with AMC.

Methods

The review protocol is available on the University of York Centre for Reviews and Dissemination database PROSPERO; registration number CRD42014013307; registration date 28/08/2014. The processing and reporting of this review are consistent with the general recommendations provided by the PRISMA revision [25].

The following online databases were searched electronically: PubMed, Medline via OvidSP, ClinicalKey, Informit and the Cochrane library. The search terms used were “acquired megacolon”, “idiopathic megacolon”, “dolichocolon” and “redundant colon”. Two independent reviewers developed inclusion and exclusion criteria.

The selection criteria were primary studies:

  1. 1)

    Diagnosing AMC using radiological, histological, laparoscopic or open surgical, endoscopic or other means;

  2. 2)

    Investigating the symptomatology and presentation of AMC; and

  3. 3)

    Providing pre-operative data.

The exclusion criteria were:

  1. 1)

    Primary studies with exclusively post-operative data;

  2. 2)

    Acquired megarectum-predominant disease;

  3. 3)

    Exclusively paediatric studies;

  4. 4)

    Studies of organic or obstructive causes of megacolon;

  5. 5)

    Animal models;

  6. 6)

    Studies that failed to exclude organic causes of megacolon;

  7. 7)

    Studies with less than three patients; and

  8. 8)

    Full text not available.

Studies published in English, from randomised controlled trials, non-randomised trials, cohort studies or case series consisting of three or more patients were selected. As there were few studies meeting these criteria no limit was put on date of publication.

Kantor (1924) failed to definitively exclude organic causes for megacolon. Despite satisfying a component of exclusion criteria, this primary publication was referenced by nearly every other study included in this review. Deemed a vital contributor to the study of AMC, both reviewers allowed the inclusion of this study in the review [7].

Factors including study design, year of publication, numbers of patients, controls and methods used to exclude organic disease were recorded. Themes relating to diagnostic criteria, colonic dimensions, histology, colonic transit time and anorectal manometry were recorded, as were patient demographics.

Mean, range, standard deviation and statistical significance were pooled and provided in the review. Conclusive findings were discussed where available. Meta-analysis were not performed due to the heterogeneity of the studies. Authors agreed to exclude individual patients with incomplete data and rectal predominant disease.

Results

The literature search identified 1205 publications of potential interest and 23 of these fulfilling inclusion and exclusion criteria (Fig. 1), described 532 patients with AMC (Table 1). A slight female preponderance was observed with a mean age of 52 years (Table 2).

Fig. 1
figure 1

PRISMA 2009 flow diagram

Full size image
Table 1 Characteristics of Studies Investigating the Diagnosis of AMC
Full size table
Table 2 Age and Sex Distribution of Patients Studied with AMC
Full size table

Exclusion of organic disease

All studies, with the exception of Kantor (1924), excluded organic disease by demonstrating an intact anorectal inhibitory reflex, the absence of hypoganglionosis on rectal biopsy or a combination of the two. Two studies showed an abnormal anorectal inhibitory reflex but normal histology [19, 20]. The anorectal inhibitory reflex was present in all patients in eight studies [3, 9, 19, 22,23,24, 26, 27]. Three of four patients of Yoshino et al. (2007) described intact anorectal inhibitory reflexes [28]. Twenty-one of 26 patients of Gattuso et al. (1997) had an intact anorectal inhibitory reflex [22]. One study in an endemic area for trypanosoma Cruzii performed three consecutive pathology screens to exclude Chagas Disease [24].

Physical dimensions

A variety of studies recorded colonic diameters of patients with AMC or provided dimensions used to diagnose its presence (Table 3). Recent contributions on this topic emphasise colonic diameter as a defining feature [9, 19, 26, 27, 29,30,31,32]. Earlier studies referred to length [7, 30, 31, 33]. Twelve studies used barium enema or similar studies [3, 7, 9, 19, 27, 29, 30, 32, 34, 35]. Taylor et al. (1980) was the only study to use abdominal X-rays [26].

Table 3 Imaging Findings Used to Diagnose AMC
Full size table

Using 50 double contrast barium enema studies, Preston et al. (1985) concluded a normal diameter of the rectosigmoid at the pelvic brim was less than 6.5 cm [9]. Six studies of this review used this criterion to diagnose AMC [23, 27, 32,33,34,35]. The three studies, Preston et al. (1985), Iantorno et al. (2007) and Koch et al. (1993) calculated mean sigmoid diameters of 10+/− 3.5 cm, 10+/− 2 cm and 10 cm respectively for sigmoid-inclusive AMC [9, 24, 35].

The presence of colonic reduplications, angulations or loops seen during barium studies or enema-filled sigmoid loops rising above the iliac crests was described by Kantor (1924) [7]. Taylor et al. (1980) diagnosed AMC by an increased colonic diameter on radiological imaging and chronic constipation [26]. Ryan (1982) described the recurrence of both colonic dilatation and symptoms in patients who had undergone segmental colectomy for megacolon volvulus of previously non-dilated segments [29].

Kantor (1924) demonstrated the limitations of simple X-ray techniques, with individual colonic dimensions varying significantly with serial imaging amongst individuals [7].

Symptoms

Ten studies investigated symptoms associated with AMC, for which eight are tabulated [3, 7, 19, 21,22,23,24, 27] (Table 4). Most predominant symptoms were constipation, abdominal pain, distention and gas distress.

Table 4 Occurrence of Symptoms in Patients with a Diagnosis of AMC
Full size table

Both Lane and Todd (1977) and Stabile et al. (1991) reported adults presenting with constipation, distension and abdominal pain while children presented with faecal impaction and soiling [3, 19]. The study by Barnes et al. (1986) described both patients with AMC and acquired-megarectum (AMR) with distal colonic involvement. AMC-specific data could not be clearly extrapolated from those with AMR. Therefore, it could not be included in the final data tabulation. This study reported symptoms in children with early onset (i.e. less than 10 years old) as faecal soiling, constipation, distension, abdominal pain, rectal impaction and abdominal mass on palpation. Patients with onset of symptoms later in life (i.e. greater than 10 years old) had constipation, distension and abdominal pain [20]. Kantor (1924) estimated that 23% of patients with constipation have AMC [7].

Colonic histology

Histological findings were presented in 11 papers [23, 24, 27, 30, 32, 34,35,36,37,38,39] (Table 5). As different techniques, sample sizes and sites were used, no definite conclusions could be drawn. A general theme of enteric architectural and neurochemical abnormality in patients with AMC was inferred.

Table 5 Histopathological Findings of AMC
Full size table

Three studies demonstrated decreased concentration of interstitial cells of Cajal, [23, 24, 39] three suggesting diminished ganglia, [30, 38, 39] four with diminished enteric neural densities [23, 27, 35, 38] and three suggesting enteric smooth muscle hypertrophy [27, 32, 35]. Conversely, other studies showed normal enteric neuron histology and enteric muscle thickness [3, 19, 20].

Koch et al. (1993) and Koch et al. (1996) investigated colonic wall neurotransmitters and enzyme systems in AMC. Conclusions made from these studies include decreased vasoactive intestinal peptide (VIP) and acetylcholinestrase (AChE) activity in the muscularis externa and increased nicotinamide adenine dinucleotide phosphate (NADPH) diaphorase activity in the muscularis externa [34, 35]. Koch et al. (1993) suggested a hypothesis for the development of AMC involving: (1) colonic hypertrophy resulting from prolonged cholingeric nerve mediated contractions of the circular muscularis externa; or (2) colonic dilatation secondary to prolonged contraction of the longitudinal smooth muscle [35].

Okhubo et al. (2014) demonstrated that non-dilated colonic loops exhibited similar histopathological abnormalities as dilated loops in AMC and that histopathological abnormalities preceded clinical symptoms in some circumstances [39]. Genomic sequencing by Chen et al. (2002) found no mutation of neurturin [33].

Anorectal manometry

Techniques used to study anorectal pressures varied. One study found resting anal canal pressures in AMC were higher than controls [26]. Half of patients of Taylor et al. (1980) recorded anal canal pressures higher among AMC than controls [26]. Yoshino et al. (2007) found patients with AMC had a higher incidence of very slow anal pressure waves [28].

Rectal sensation was decreased in 50% and normal in 50% of patients with AMC in the study by Lane and Todd study (1977). Increased rectal capacity was also noted [19]. Chen et al. (2002) described rectal hyposensation in four of five AMC cases [33]. Diminished rectal sensitivity to balloon distension but intact perianal sensation and rectal electrosensation was described by Koch et al. (1997) [22].

Colonic transit time

Prolonged colonic transit times were recorded in majority of patients by Chen et al. (2002) and Stabile et al. (1991) [3, 33]. The study by Yoshino et al. (2007) demonstrated very slow periodical pressure changes in the anal canal signifying ultra-slow colonic transit time (OR 10.67, 95% CI 4.40–25.86) [28].

Complications

Five studies reported incidence of volvulus, colonic obstruction or faecal impaction with AMC [22, 30, 31, 39, 40]. Sigmoid volvulus occurred as follows: Kobak et al. (1962) two patients; Rios-Dalenz et al. (1975) 49 patients; Ohkubo et al. (2014) two patients and Gattuso et al. (1997) five patients [22, 30, 31, 39]. Goulston (1976) described a 6.6:1 rate of impacted faeces, megacolon and volvulus among a psychiatric hospital patients in comparison to a general hospital [40].

Neuropsychiatric conditions

Six studies showed an association of AMC with neurological disease [3, 28, 30, 34, 35, 40]. Goulston et al. (1976) reported faecal impaction, volvulus and megacolon were higher in a psychiatric hospital, in comparison to a general hospital, 0.53% to 0.08% respectively [40].

Three studies recorded psychiatric conditions, predominantly schizophrenia, among their patients [30, 34, 35]. Cerebrovascular accidents and other organic central neurological conditions such as epilepsy were also of note [3, 28, 30, 34, 35]. Two patients of Gattuso et al. (1997) suffered mental retardation [22].

Discussion

This review sought to evaluate the diagnosis of AMC. Twenty-three studies were identified, 17 of which had control patients. The sample size in many of these studies was small and few studies provided the statistical significance of their findings. Conclusions drawn are inevitably tentative.

Demographics

Patients with AMC may present at in both genders and at any age, though the presentation of children differ. There may be an association with neuro-psychiatric conditions and medications used in this patient group. It is unknown if the condition is a result of inappropriate behavioural response to defaecation, enteric physiological impairment or is associated with the use of medication.

Symptoms

The common features of an adult presentation were constipation, distension, gas distress and abdominal pain. In comparison, children presented with faecal incontinence and impaction.

No study assessed the impact on quality of life. There may well be a considerable overlap between the symptoms associated with AMC and Constipation Predominant Irritable Bowel Syndrome given 20% of constipated patients have an AMC [7]. Brummer et al. (1962) supports this, estimating that 30% of patients with constipation have an AMC [16]. Whether patients with AMC are being misdiagnosed as having Constipation Predominant IBS cannot be deduced from this review.

Imaging

Imaging, usually barium enema, has been used. The bowel appears as a grossly elongated and dilated colon with multiple loops [3, 9, 19, 24, 29, 35,36,37]. The Preston et al. (1985) study, referenced by six studies in this review, proposes that rectosigmoid diameter at the pelvic brim exceeding 6.5 cm as diagnostic [9, 23, 27, 32,33,34,35]. Gladman et al. (2007) reported weaknesses with this study involving their choice of control group, despite its use by many later reports [1]. A mean sigmoid diameter around 10 cm is suggested. [9, 24, 35] The sigmoid is often involved radiologically, [9, 19, 24, 29, 30, 35] but AMC may exclusively affect the proximal colon, [3, 9, 29, 41] for which this definition is not suitable.

Simple 2D imaging has limitations in assessing colonic diameters and variations occur between serial images [7]. There seems little doubt that the increasing availability of CT colonography using standardised insufflation pressure will optimise the diagnosis of AMC [42]. CT colonography has the advantage of being a quantitative imaging modality and offers an alternative approach to evaluate the colon and rectum following incomplete colonoscopy [43, 44]. Low radiation dose imaging is also possible with modern scanners [45]. CT Colonography allows simple measurement of colonic diameters and length from multiple views, has shorter procedure times, does not require recovery supervision and carries less procedural risks than traditional colonoscopy [42, 44, 46,47,48,49]. It currently has a prominent role following incomplete colonoscopy - a common occurrence amongst patients with AMC. The use of this modality is limited in screening as biopsy and polypectomy cannot be performed [44].

Colonoscopy

Many colonoscopists seem confident about diagnosing AMC or colonic redundancy, although no objective criteria for diagnosis have been defined [46, 50, 51]. This modality, does however, depend upon subjective interpretation. AMC or colonic redundancy have been associated with incomplete colonoscopy, as a result, may not be the most appropriate investigation for this population group [46, 52]. Hanson et al. (2007), analysed the colonic length of patients with redundant colon during colonoscopy using CT colonography. This study reported patients with incomplete colonoscopy as having colonic lengths exceeding 200 cm, often reported as redundant colon during colonoscopy [46]. Colonic redundancy was defined as elongated and tortuous colons or those with two or more acute flexures [46, 53]. Although the diagnosis of AMC was not pursued in this study, it may suggest that CT colonography is a useful modality to diagnose abnormal colonic dimensions, both in terms of length, diameter and possibly volume [46].

Pathophysiology

While no definitive consensus of histological or neurochemical changes was achieved, the study by Koch et al. (1993) suggests plausible mechanisms for the development of AMC and warrant further investigation [35]. Consecutive studies in this review conclude with findings of altered neurochemical and enteric architectural findings [34, 35].

AMC is a disease of exclusion. Ruling out an organic cause for this condition is pertinent. The absence of hypogangliosis has more consistent results when compared with anorectal reflex testing, although it does carry more risk.

Complications

The incidence of colonic volvulus in AMC may be underestimated in this review, as colectomy studies were largely excluded. Emergency colectomy is indicated for patients presenting acutely with volvulus as a consequence of AMC. In the elective setting, surgeons may encounter patients with AMC experiencing intractable symptoms and poor quality of life that have failed conservative management. For the surgeon, the decision to operate must be balanced against the risk of not optimising their symptoms. Amongst this demographic, elective colectomy is sometimes offered as a final attempt.

Conclusion

Based on the data collated in this review, we propose the following criteria for the diagnosis of acquired megacolon: (1) the exclusion of organic disease by rectal biopsy or an intact anorectal inhibitory reflex; (2) a sigmoid diameter of ~ 10 cm on abdominal X-ray or barium enema; (3) and symptoms including constipation, distension, abdominal pain and gas distress. Given the limitations of 2D radiological imaging, CT colonography may be a more optimal imaging modality.

The condition affects both sexes and has preponderance for the middle age [3, 7, 9, 19, 20, 22,23,24, 26,27,28,29,30,31,32,33,34,35,36,37,38,39,40]. Histologically, evidence supports enteric architectural and neurochemical abnormalities, however, detailed findings are variable [23, 24, 27, 30, 32, 34,35,36,37,38,39]. A proportion of patients with AMC may experience colonic volvulus [22, 30, 31, 39, 40]. Furthermore, the condition may have an increased prevalence among patients with neuropsychiatric conditions [3, 28, 30, 34, 35, 40].

Whether AMC is a single entity or a group of heterogeneous conditions is unknown. Neither is its relationship to other constipation predominant conditions. It may well be that patients with AMC are misdiagnosed as having Constipation Predominant IBS. The natural history of this condition and optimal forms of management are yet to be elucidated. Surgical procedures are performed on patients with AMC for intractable disease and emergency situations, risking morbidity and mortality [1, 4]. This systematic review may help in the understanding of the presentation, methods of diagnosis and some of the associations of AMC. Further research is required on the pathophysiology of the condition, protocols for conservative treatment and the place of surgery for intractable disease.