Constipation is a frequent health concern for older people in every health care setting. Primary care visits for constipation increase markedly in people older than 60 years, as does regular use of laxatives. Self-reported constipation in older people is associated with anxiety, depression, and poor health perception, while clinical constipation in vulnerable individuals may lead to complications such as fecal impaction, overflow incontinence, sigmoid volvulus, and urinary retention. Constipation is an expensive condition, with high costs ranging from laxative expenditure to nursing time. For instance, it is estimated that 80% of community nurses working with older people in the United Kingdom are managing constipation (particularly fecal impaction) as part of their case-load. An Australian study used in-depth, semistructured interviews to explore older individuals’ experiences with constipation, and their findings largely summed up feelings and problems, no doubt, shared by many older people across the developed world:

• They feel “not right” in themselves when they are constipated.
  
• Physicians can have a dismissive attitude about constipation and do not consider the problem seriously.
  
• Patients are keen to find a solution, but feel useful and empathic advice and information are generally unavailable.
  
• At the same time, they have a strong imperative for self-management including use of over-the-counter laxatives.
  
• There are some barriers to lifestyle approaches, for example, expense of fruit and vegetables, fear of urinary incontinence with increased fluid intake, reluctance to walk out alone.
  
• One-quarter still need to do self-manual removal despite measures taken.

This chapter will describe the epidemiology, risk factors, clinical presentation, assessment, and treatment of constipation in older adults. Data sources were a computer search of the English language literature (1966 to 2006), systematic review Web sites including the Cochrane database, reference lists from recent systematic reviews and book chapters, and expert committee reports and opinion. Levels of evidence are as used by the U.S. Preventive Task Force:

• Good evidence Level [1]: consistent results from well-designed, well-conducted studies
  
• Fair evidence Level [2]: results show benefit, but strength limited by number, quality, or consistency of studies
  
• Poor evidence Level [3]: insufficient because of limited number, power, or quality of studies

Definitions of constipation in older people in medical and nursing literature have been inconsistent. Studies of older people have tended to define constipation

• subjectively by self-report,
  
• according to specific bowel-related symptoms, or
  
• by daily laxative usage.

Few use objective assessment–based definitions (e.g., fecal loading). The feeling of being constipated frequently means different things to different individuals. While the nonspecific self-reporting of constipation (“I suffer from constipation”) provides insight into how individuals perceive their bowel habit, standardized definitions based on specific symptoms (Rome II criteria) are now widely used in both clinical practice and research (Table 93-1). A recent systematic review reported that approximately 63 million people in North America meet the Rome II criteria for constipation with a disproportionate number being older than 65. An important subtype of constipation in older people is rectal outlet delay, which affects 21% of community dwellers aged 65+. The Rome II definitions for constipation and rectal outlet delay are symptom-based; objectively, however, the clinical definition of constipation relies on finding fecal loading in the rectum and/or colon. Such objective assessment is particularly important in frail older people owing to factors listed in Table 93-2.

Table 93-3 provides practice guidance based on evidence from epidemiological studies (prevalence, symptomatology, and risk factors) of constipation in older people.

Self-Reported Constipation

One older community-based study of 3166 persons aged 65 years and older asked the question, “Do you have recurrent constipation?” and found a prevalence of 26% in women and 16% in men; in the 84+ years age group, prevalence was 34% and 26%, respectively. Age was a strong independent risk factor for self-reported constipation. Other community studies support this relationship with age and show prevalence rates of up to 34% of women and 30% of men older than age 65 years. The preponderance of women over men reporting constipation tends to equalize after the age of 80 years.

Infrequent Bowel Movements

Two or fewer bowel movements per week are below normal range and tend to signify slow transit constipation. Weekly frequency of bowel movements does not, however, alter with age alone, in contrast to self-reporting of constipation. In community-based studies, it has been found that

• only 1% to 7% of both younger and older community-dwelling individuals report two or fewer bowel movements a week;
  
• this consistent bowel pattern across age groups persists even after statistical adjustment for the greater amount of laxatives used by older people;
  
• among older people complaining of constipation, less than 10% report two or fewer weekly bowel movements, and more than 50% move their bowels daily.

Difficult Evacuation

So, what are the symptoms other than infrequent bowel movements that drive self-reporting of constipation in older people? Theses symptoms are predominantly straining and passage of hard stools. Of older people reporting constipation in a U.S. community study, 65% had persistent straining and 39% had passage of hard bowel movements. Difficult rectal evacuation is a primary cause of constipation in older people. Twenty-one percent of community-dwelling people aged 65+ had rectal outlet delay (according to Rome II criteria), and many describe the need to self-evacuate. Among frailer individuals, difficult evacuation can lead to rectal impaction and fecal soiling.

Constipation Symptoms in the Long-Term Care Setting

Long-term care residents are at increased risk of developing complications of constipation (Table 93-4) that may precipitate acute hospital admissions. Physical frailty in older persons does increase the prevalence of infrequent bowel movements, with 17% of nursing home residents and 14% of geriatric day-hospital attendees reporting two or fewer bowel movements a week. Among long-term care residents self-reporting constipation, 33% have two or fewer bowel movements a week. A Finnish study showed the prevalence of chronic constipation and/or rectal outlet delay to be 57% in women and 64% in men living in residential homes, and 79% and 81% respectively in the nursing home setting. A U.K. study found that 64% of nursing home residents taking laxatives still reported straining on more than 1 in 4 occasions. This and the fact that 50% to 74% of long-term care residents use daily laxatives suggest that rectal evacuation difficulties are not being well managed in this population.

Physiological studies suggest that changes in the lower bowel predisposing toward constipation in older people are not primarily age-related. This is compatible with the epidemiology showing that (1) bowel movement frequency does alter with aging, and (2) constipation symptoms are more prevalent in older people with comorbidities. Extrinsic causes such as reduced mobility, fluid intake, dietary fiber, comorbidities, and medication all impact colonic motility and transit, and influence the pathophysiology of constipation.

Colonic Function

Colonic motility depends on the integrity of the central and autonomic nervous systems, gut wall innervation and receptors, circular smooth muscle, and gastrointestinal hormones. Propagating motor complexes in the colon are stimulated by increased intraluminal pressure generated by bulky fecal content. Studies of total gut transit time (passage of radiopaque markers from mouth to anus, normally 80% passed within 5 days), colonic motor activity, and postprandial gastrocolic reflex show no differences between healthy older and younger people. Older people with chronic constipation do, however, tend to have a prolonged total gut transit time, ranging from 4 to 9 days. Radiologic markers pass especially slowly through the left colon with striking delay in the rectosigmoid, suggesting that total transit time is prolonged because of segmental dysmotility in the “hindgut.” The prolongation in transit time is even greater in institutionalized or bedridden patients with constipation, with total gut transit time ranging from 6 to more than 14 days. Slow transit results in a cycle of worsening colonic dysfunction by reducing water content of stool (normally 75%) and shrinking fecal bulk, which then diminishes the intraluminal pressures, and hence the generation of propagating motor complexes and propulsive activity.

Intrinsic Mechanisms for Colonic Dysfunction in Older People with Constipation

Certain intrinsic mechanisms for altered colonic function in older persons with constipation have been postulated from physiologic studies (Table 93-5). Overall collagen deposition in the left side of the colon increases with aging, and this could alter colonic compliance and motility. Direct electrophysiologic measurement of colonic motor activity in elderly subjects has shown that the sigmoid motor response to intraluminal Bisacodyl (a direct stimulant of the myenteric plexus) is diminished in patients who are constipated, implying a deficit in intrinsic innervation. Myenteric plexus dysfunction may partially account for impaired gut motility in elderly persons with constipation. The total number of neurons in the myenteric plexus is decreased, and this neuronal loss bears no relation to the presence of pseudomelanosis coli cells, implying that use of anthraquinone laxatives is not the primary cause.

Another possible intrinsic factor is age-related deficit in the density of inhibitory nerves, or in the binding sites on smooth muscle for inhibitory gut neuropeptides. In vitro studies of colons across age groups showed an age-related reduction in the amplitude of inhibitory junction potentials, but no decrease in the levels of inhibitory gut neuropeptides. This age-related decline occurs earlier in women as compared with men. Such a decrease in inhibitory nerve input to the circular smooth muscle could result in segmental motor incoordination, which may lengthen transit time and promote constipation in older persons with other predisposing risk factors. Individuals older than age 60 years have higher plasma concentrations of beta-endorphin with increased binding to opiate receptors in the gut wall and myenteric plexus. Higher opiate binding has the effect of relaxing colonic tone, reducing motility, and inhibiting the gastrocolic reflex.

Finally, it is interesting to note that constipation is more prevalent in patients with nonulcer dyspepsia, the two clinical conditions being linked by gastrointestinal hypomotility.

Anorectal Function

In normal defecation, colonic activity propels stool into the rectal ampulla causing distension and intrinsically mediated relaxation of the smooth muscle of the internal anal sphincter (or anal canal). This is followed promptly by reflex contraction of the external anal sphincter and pelvic floor muscles, which are skeletal muscles innervated by the pudendal nerve. The brain registers a desire to defecate, the external sphincter is voluntarily relaxed, and the rectum is evacuated with assistance from abdominal wall muscle contraction.

Age-Related Changes in Anorectal Function

There is a tendency toward an age-related decline in internal sphincter tone, particularly in the eighth decade onward. Clinically, this predisposes older individuals to fecal incontinence, particularly with loose stools. There is a more definite age-related decline (greater in women than men) in external anal sphincter and pelvic muscle strength, which can contribute toward evacuation difficulties. Failure of the anorectal angle to open and excessive perineal descent in older women can lead to constipation. In simulated defecation studies, 37% of nonconstipated older subjects were unable to evacuate a small solid sphere. Consequent prolonged straining may compress the pudendal nerve, further exacerbating any preexisting weakness. There appears to be a reduction in rectal motility with normal aging, again in older old age. Rectal sensation does not alter with normal aging.

Anorectal Dysfunction in Older Persons

There are three types of anorectal dysfunction that predispose older people to rectal outlet delay.

The most common is rectal dysmotility characterized by reduced rectal motility, increased rectal compliance with a variable degree of rectal dilatation, and impaired rectal sensation such that the urge to pass stool is blunted. Over time, an increasing degree of rectal distension is required to reflexly trigger the defecation mechanism. These patients have rectal retention of hard or soft stool on digital examination of which they may be unaware. The resulting rectal distension leads to relaxation of the internal sphincter and hence to fecal soiling. One study showed that rectal contractions could be elicited in only 14% of older people with a history of rectal impaction. One postulated cause for rectal dysmotility is diminished parasympathetic outflow as a result of impaired sacral cord function, for example, from ischemia or spinal stenosis (Table 93-5). In a significant number of older people, rectal dysmotility can develop through a persistent disregard or suppression of the urge to defecate as a result of dementia, depression, immobility, or painful anorectal conditions. Voluntary increase in intra-abdominal pressure during defecation could overcome rectal dysmotility to produce enough of an increase in rectal pressure for evacuation to occur, but older people often have weakened abdominal musculature, limiting their ability to compensate in this way.

Pelvic floor dyssynergia, though more common in younger women, can cause rectal outlet delay in older people. This is caused by paradoxical contraction or failure to relax the pelvic floor and external anal sphincter muscles during defecation, and manometric studies show paradoxical increases in anal canal pressure on straining. This abnormal expulsion pattern may be seen in individuals with severe and long-standing symptoms of rectal outlet delay, and in patients with Parkinson disease.

Another type of anorectal dysfunction is irritable bowel syndrome (IBS) characterized by increased rectal tone and reduced compliance. Sensation of pain on distending the rectum during anorectal function tests has been shown to be greater in patients with IBS than controls. IBS is usually constipation-predominant in older people. These patients are likely to have a many year history of difficult passage of small fecal pellets and other IBS symptoms such as passing mucus, abdominal distension, and pain.

Both the epidemiology and pathophysiology of constipation in older people point to the enormous importance of identifying predisposing causes for the condition in each affected individual. One prospective study examined baseline characteristics predictive of new-onset constipation in elderly nursing home patients, using the U.S. Minimum Data Set instrument. Seven percent (n = 1291) developed constipation over a 3-month period. Independent predictors were white race, poor consumption of fluids, pneumonia, Parkinson disease, allergies, decreased bed mobility, arthritis, greater than five medications, dementia, hypothyroidism, and hypertension. The authors postulated that allergies, arthritis, and hypertension were associated primarily because of the constipating effect of drugs used to treat these conditions. Other studies have shown that institutionalization itself is an independent risk factor for symptom-based constipation in older people. Table 93-6 summarizes evidence-based risk factors of constipation in the elderly population.

Reduced Mobility

Impaired mobility is a common risk factor for constipation in older people. Greater physical activity (including regular walking) is associated with less self-reported and symptom-specific constipation in older people living both at home and in long-term care. Reduced mobility was found to be the strongest independent correlate of heavy laxative use among nursing home residents, following adjustment for age, comorbidity, and other relevant clinical factors. Gut transit time in elderly subjects was measured independently as 3 days in ambulant, and 3 weeks in bedridden patients, although comorbid factors were likely to be contributory. A study of healthy young male volunteers showed that after only 1 week of bed rest, both transit through the sigmoid colon and stool frequency were reduced. It is well documented that exercise increases colonic propulsive activity (“joggers diarrhea”), especially when measured postprandially. In a population survey of younger women (36–61 years), daily physical activity was associated with less constipation (defined as two or fewer bowel movements per week), and the association strengthened with increased frequency of physical activity. This leads to speculation that increasing physical activity in adulthood may reduce the likelihood of constipation problems in older age.

Polypharmacy/Drug Side Effects

Polypharmacy itself increases the risk of constipation in older patients, particularly in nursing homes where each individual takes an average of six prescribed medications per day.

Anticholinergic medications reduce contractility of the smooth muscle of the gut via an antimuscarinic effect at acetylcholine receptor sites, and in some cases (e.g., patients with schizophrenia taking neuroleptics), long-term use may result in chronic megacolon. In two cross-sectional studies of nursing home residents, anticholinergic antidepressants were independently associated with daily laxative use following adjustment for age, gender, function, and cognition. Anticholinergic neuroleptics and antihistamines were also independently associated in one of the studies; nonanticholinergic sedatives, however, were not found to be constipating. A recent study of 532 community-dwelling older U.S. veterans found that among the 27% using anticholinergic drugs, the rate of constipation (42%) was significantly greater than those not using the drugs.

While older people are very susceptible to the constipating effects of opiate analgesia, a recent U.S. study of nursing home residents with persistent nonmalignant pain found that there was no increased rate of constipation in chronic opiate users over a 6 month period compared to those not taking opiates. They also observed a general improvement in functional status and social engagement. Constipation in chronic opiate users can be effectively managed (by laxative or suppository coprescription where needed)—an important finding as chronic pain is often undertreated in frailer older people perhaps owing to fear of the adverse effects of analgesic drugs. In terms of different preparations, community-based studies of adults receiving opiates for chronic pain have shown equal constipation risk for all sustained-release oral preparations. Transdermal patches (e.g., fentanyl), however, are associated with lower risk of constipation than oral preparations.

All types of iron supplements (sulphate, fumarate, and gluconate) cause constipation in adults, the constipating factor being the amount of elemental iron absorbed. Slow-release preparations have a lesser impact on the large bowel, but this is because they tend to carry the iron past the first part of the duodenum into an area of the gut where elemental iron absorption is poorer. Administration of iron sulfate in doses greater than 325 mg per day does not substantially increase iron absorption in elderly persons and may significantly increase gastrointestinal side effects. Intravenous iron does not cause constipation and may be an alternative in patients with chronic anemia (e.g., chronic kidney disease) who have symptomatic constipation on oral iron.

In a recent 5-year study of calcium supplementation in older women, the only side effect was constipation (treatment 13.4% vs. placebo 9.1%). The study showed that calcium supplementation reduced bone loss and turnover and fracture rates in older women who took it, but long-term compliance was poor, and constipation may have contributed to this.

Calcium channel antagonists impair lower gut motility, particularly in the rectosigmoid, by inhibiting calcium uptake into smooth muscle cells and altering intraluminal electrolyte and water transportation. Severe constipation has been reported in older patients taking calcium channel antagonists, with nifedipine and verapamil being the most potent inhibitors of gut motility in this class of drugs.

Nonsteroidal anti-inflammatory drugs (NSAIDs) increase the risk of constipation in older people, most likely through prostaglandin inhibition. In a large case-controlled primary care study, constipation and straining was a more common reason for stopping NSAIDs than dyspepsia. NSAIDs have also been implicated in causing stercoral perforation in patients with chronic constipation. Using 15 or more aspirin tablets a week has been linked to constipation in a middle-aged cohort of women.

Aluminium antacids have been associated with constipation in older people living in both nursing homes and in the community.

Dietary Factors

Fiber

Low consumption of wheat bran, fiber, vegetables, fruit, rice, and calories can all predispose toward constipation. A U.K. survey showed that consumption of fruit, vegetables, and bread decreases with advancing age. It has been suggested that the prevalence of constipation is rising because modern food processing produces refined food with low roughage. Community studies of older Europeans who eat a Mediterranean diet rich in fruit, vegetables, and olive oil show a low prevalence of constipation (4.4% in people aged 50+). Conversely, a German questionnaire survey of adults with and without constipation reported that chocolate, white bread, and bananas were the foodstuffs most strongly perceived to harden stools.

Calories

Low calorie intake in older people (adjusted for fiber intake) has been linked to constipation. One study looked at nutritional factors across all nursing homes in Finland and found that malnutrition and constipation were associated. This may be a two-way association in that marked constipation or fecal impaction can cause anorexia, while low calorie intake can promote constipation.

Enteral Nutrition

Constipation is a recognized problem in patients receiving enteral nutrition. A prospective survey from Spain of hospitalized patients (mean age 76) receiving nasogastric tube feeding identified constipation as a complication of treatment in 30%. Enteric feeding products containing fiber are available, though there are no data on whether constipation is any less of a problem with their use.

Fluid Intake

Amount

Low fluid intake in older adults has been related to symptomatic constipation in epidemiologic surveys and (in an unadjusted analysis) to slow colonic transit. In patients with Parkinson disease, low water intake correlated with severity of constipation. Withholding fluids over a 1-week period in young male volunteers significantly reduced stool output. Elderly people are at greater risk of dehydration because of

• impaired thirst sensation,
  
• less effective hormonal responses to hypertonicity,
  
• limited access to drinks because of coexisting physical or cognitive impairments,
  
• voluntary fluid restriction in an attempt to control urinary incontinence.

Alcohol and Coffee

A large Japanese survey of constipation symptoms found that alcohol consumption was a preventive factor in men. A population survey of middle-aged women in the United States showed that daily alcohol consumption (exceeding 12 g/d) and low-moderate caffeine intake were independently inversely related to infrequent bowel movements. Black coffee has been shown to increase colonic motility specifically in the rectosigmoid within 4 minutes of ingestion in young healthy volunteers (a reaction not observed with ingestion of hot water), implying that caffeine triggers the gastrocolic reflex.

Parkinson Disease

Patients with Parkinson disease suffer from three primary pathologies that lead to constipation:

• Primary degeneration of dopaminergic neurons in the myenteric plexus resulting in prolonged colorectal transit
  
• Pelvic dyssynergia causing rectal outlet delay and prolonged straining
  
• Small increases in intra-abdominal pressures on straining (compared with age-matched controls)

Constipation can become prominent early in the course of the disease, even 10 to 20 years prior to motor symptoms. In a 24-year longitudinal study (Honolulu), less than one bowel movement a day was associated with a threefold elevated risk of future Parkinson disease in men. A recent study of patients at a Parkinson disease clinic found that 59% were constipated according to the Rome criteria (vs. 21% in age-matched control group without neurological disease), and 33% were very concerned by their bowel problem. Antiparkinsonian drugs can further exacerbate constipation. Pelvic dyssynergia affects 60% of people with Parkinson disease and may be hard to treat. Botulinum toxin injected into the puborectalis muscle has been used to improve rectal emptying in Parkinson disease patients with good effect, though repeat injections every 3 months are required to maintain clinical benefit.

Diabetes Mellitus

A Turkish study of outpatients with type 2 diabetes showed that 56% complained of constipation (vs. 30% of controls). Neuropathy symptom scores correlated with laxative usage and straining. Diabetic patients with autonomic neuropathy are more likely to be constipated because of markedly slowed transit throughout the colon and impairment of the gastrocolic reflex. However, one-third of diabetic patients with constipation do not have neuropathic symptoms, so additional potentially reversible factors should be considered particularly in older people (e.g., drugs, mobility, fluids). Indeed, a U.S. community study found that constipation and/or laxative use was increased in type 1 versus type 2 diabetic men, but this difference was associated with use of calcium channel-blockers rather than with neuropathy symptoms. Acute hyperglycemia inhibits the gastrocolic reflex and colonic peristalsis, so glycemic control is important. Colonic transit time in frail and immobile older people with diabetes is extremely prolonged at 200 ±144 hours. An Israeli study showed that this very long transit time in long-term care residents with diabetes can be significantly reduced by administering acarbose, an alpha-glucosidase inhibitor with a potential adverse effect of causing diarrhea. Overall, gut dysmotility can lead to bacterial overgrowth and the clinical problem of explosive diarrhea; treatment with erythromycin and long-term motility agents such as metoclopramide should be considered in these individuals.

Dementia

Dementia predisposes individuals to rectal dysmotility, partly through ignoring the urge to defecate. A study in which young men deliberately suppressed defecation resulted in prolonged transit through the rectosigmoid with a marked reduction in frequency of bowel movements. Epidemiological studies show a significant association between cognitive impairment and nurse-documented constipation in nursing home residents. Patients with non-Alzheimer dementias (Parkinson disease, Lewy body, vascular dementia) compared to those with Alzheimer dementia are more likely to suffer from autonomic symptoms, including constipation.

Mood-Related Disorders

Depression, psychological distress, and anxiety are all associated with increased self-reporting of constipation in older persons. In certain cases, the symptom of constipation is a somatic manifestation of psychiatric illness. A careful assessment is required to differentiate subjective complaints from clinical constipation in depressed or anxious patients.

Stroke

Constipation affects 60% of those recovering from stroke on rehabilitation wards, and a high number of these have combined rectal outlet delay and slow transit constipation. For stroke survivors living in the community, problems relating to bowel evacuation are greatly worsened by difficulties accessing the toilet owing to functional impairment. Fecal incontinence in stroke survivors has been shown to relate more to modifiable disability-related factors such as toilet access and anticholinergic medication use than to stroke-related factors (such as severity and lesion location). Weakness of abdominal and pelvic muscles following stroke also contribute to problems with evacuation.

Spinal Cord Injury/Disease

Constipation affects the majority of people with spinal cord disease or injury. Age and duration of injury interact to promote complications of chronic constipation such as acquired megacolon, which affects more than half of patients with spinal cord injury. Lumbar stenosis in older people caused by degenerative joint disease may lead to cauda equina problems with severe rectal outlet delay. One study in younger people showed that an average of 27% (range 0–44%) of rectosigmoid emptying was achieved with each defecation in patients with cauda equina syndromes, versus 81% (range 53–100%) in healthy controls.

Metabolic Disorders

Hypokalemia produces neuronal dysfunction that minimizes acetylcholine stimulation of gut smooth muscle and so prolongs transit through the gut. It should be excluded in cases of colonic psuedo-obstruction and sigmoid volvulus. Hypercalcemia causes conduction delay within the extrinsic and intrinsic innervation of the gut. Surgical treatment of hyperparathyroidism reverses the neuromuscular bowel dysfunction seen with this condition. Patients with myxedema have been observed to have edema of the gut wall with mucopolysaccharide deposition, although whether this contributes to the colonic hypomotility seen commonly in clinical hypothyroidism is uncertain. Patients on long-term renal dialysis have prolonged age-adjusted transit time, more so in hemo- than peritoneal dialysis. In a questionnaire study from Japan, 63% of hemodialysis patients complained of constipation. Important contributors to this problem were thought to be high (49%) use of resin to avoid hyperkalemia, suppression of the defecation urge while undergoing dialysis, and low fiber intake. Resin administration also places elderly inpatients at risk of fecal impaction.