Chronic constipation (CC) is associated with a number of serious gastrointestinal (GI) disorders, including colorectal cancer (CRC), gastric cancer and ischemic colitis (IC) in younger patients under the age of 50, according to a new study reported in the American College of Gastroenterology (AGS) Annual meeting 2015.1
“Younger patients presenting CC symptoms warrant consideration for colonoscopy or abdominal imaging screening, even though in the absence of red flag signs such as anaemia, bleeding or persistent weight loss”, believed Dr Lauren Gerson, MD, the study lead author from the California Pacific Medical Center, US. This is in contrast to AGS’s negative recommendation for colonoscopy screening for age<50 except presenting alarming symptoms, she acknowledged in AGS 2015 press brief2.
CC is a common functional gastrointestinal disorder in the community affecting approximately 14% of the different population, according to a pooled analysis of 41 separate study populations containing 261,040 subjects.3 While more common in elderly females, CC affects all age groups. In Hong Kong, CC, based on stringent Rome II criteria, has a prevalence of 14% and does not differ between the younger and the older people.4
Dr. Lauren and her colleagues noticed that while CC is prevalent in younger population (age<50), whether it is a risk factor for associated severe GI complications such as cancers have not been well characterized. They conducted a retrospective case-control analysis of a large electronic health records (Humedica with 6 million subjects) for the severe GI events in patients with CC in US. 12,838 subjects aged ≤50 with CC were identified and controls were propensity score matched to cases at 1:1 using age, gender, race, family GI history, comorbidities of depression and anxiety, and geographical region.1
CC was defined as at least 2 outpatient visits with an ICD-9 code of 564.0x (Constipation) at least 30 days apart, and severe GI events of GI cancers, IC, etc., were defined by ICD-9. Analysis (proportional hazards models for each GI event) were based on subjects with first GI events at least 90 days after the index date of CC to adjust for potential confounding of the outcome and initial presentation of CC.1
The median follow-up for CC cases and controls were 31 months and 26 months respectively. Mean age was 36 years and 75% of patients were female.2
In general, patients with CC showed significantly higher risk (>3 fold, HR=3.28, 95% CI 2.19‒4.90) of any serious GI disorder than the controls (no CC), with more than 5 folds increase in IC, CRC, GI cancer and diverticulitis (Table 1). CC patients also had 59% higher risk to have IBD but the difference was not significant.1,2
<<Insert Table >>
“There were 34 young cases (or absolute risk of 0.26%) of GI cancers associated with CC; half were CRC (0.14%). This is higher than the 10-year-risk of CRC for a 30 year-old based on CDC data which is 0.07%”, Dr. Lauren emphasized in the press brief, “less than 25% of these patients had bleeding and anaemia but CC only.”2
The association of some GI events as been shown before but not to this degree, she added. It was not a surprise to see the association between diverticulitis and CC, but is definitely for GI cancers, such as CRC.
Discussing this paper in the AGS 2015 press briefing session “Let’s Talk Constipation”, as moderator, Dr. Lawrence R Schiller, MD from Baylor University Medical Center in Dallas, US, pointed out that colorectal cancer is generally a slow progressing disease. He questioned if detecting GI complications, especially for cancers, 90 days after initial diagnosis of CC might be sufficient to establish CC a risk factors. “CC could be a presenting symptom of CRC as alternative interpretation”, he wondered.2
Dr. Lauren admitted she was unable to determine CC as a presenting symptom or a causal factor of the GI complications. “The use of 90 days lapse in detecting GI complications was arbitrary with aim at preventing potential confounding outcomes. We could extend the lapse to a year but could end up excluding important GI events”, she responded. But, 90 days was the minimum lapse time for GI event detection, and “the patients with CC were followed up for 31 months and controls for 27 months”, she emphasized.
Nevertheless, Dr. Schiller believed that “this study will have potential to change some of the approaches and follow-ups to the younger patients diagnosed with CC”.2
When talking about how her study could impact clinical practice, Dr. Lauren described it might not be easy to get patients describing about their poo habit, but “the use of Bristol stool chart could engage with patients for giving more details about the stool and stuff, hence facilitating the CC identification”. Once constipation is identified, it would be important to evaluate if this is a major change from the past medical history, and then see if this is chronic. Regardless of being a risk factor or a presenting symptom, CC may prompt a further colonoscopy investigation especially when the patient presents other alarming symptoms.2