Contents - Previous - Next

Impact of gastrointestinal function on protein-energy interactions and nutritional needs


* Rebecca Roubenoff died on March 12, 1992, as she was completing a graduate degree in Nutrition at Tufts University. This paper represents one of the many contributions of her short career, demonstrating her deep commitment to good nutrition and health in the world.

** UDSA Human Nutrition Research Center on Aging at Tufts University, 711 Washington Street, Boston, MA 02111, U.S.A.

1. Gastrointestinal function in protein-energy malnutrition
2. Diarrheal diseases
3. Nutritional recommendations


Protein-energy malnutrition (PEM) is associated with atrophy of the gastric mucosa and pancreas and morphological changes in the intestine. Intestinal changes are not distinguishable from morphological and functional abnormalities termed 'tropical enterophathy' since its highest prevalence is in several tropical locations lacking adequate sanitation. Clinically these changes may present as hypo- or achlorhydria, bacterial proliferation in the stomach and upper intestine and malabsorption. Acute and chronic diarrhea and tropical enteropathy, often with colonization by enteric pathogens and subclinical malabsorption, are often superimposed on PEM.

Malabsorption associated with acute, infectious diarrheas can result in the loss of 7% of yearly food energy; subclinical malabsorption nutrient losses in adults might equal 4-6% of yearly food energy. The consequences vary with the adequacy of nutrient intake and presence or absence of concurrent infection. Protein-energy interactions include: (1) at any given energy intake, increasing protein will improve nitrogen retention until physiologic needs for nitrogen balance are approached, (2) at any given protein intake, increasing energy will improve nitrogen retention, (3) the efficiency of nitrogen retention will parallel the degree of malnutrition, and (4) in the hypermetabolic phases of infection, because of increased catabolism, there will be less nitrogen retention for any given energy and protein intake. Special attention should be given to situations where caloric intake is normal but protein intake is inadequate, because this has been associated with more marked intestinal morphological changes than when both protein and caloric intakes are decreased. Outcomes in investigations designed to assess nutritional needs in clinical or subclinical malabsorption should be evaluated under usual living conditions and dietary intake to avoid erroneous conclusions regarding prevalence or nutritional consequences.

1. Gastrointestinal function in protein-energy malnutrition

Protein-energy malnutrition (PEM) is associated with significant atrophy of the gastrointestinal tract (WINICK, 1979; KEYS et al., 1950). This atrophy encompasses the gastric and intestinal mucosa and pancreas. Functionally, there is hypo- and achlorhydria, bacterial proliferation in the stomach and upper intestine, and malabsorption of carbohydrate, protein, fat, vitamin B12 and other micronutrients. This malabsorption has several etiologies (GRANT, 1983): (1) hypoproteinemia may cause bowel wall edema interfering with the passage of nutrients across the mucosal membrane, (2) decreased absorptive surface and decreased brush-border enzyme concentrations diminish absorption, (3) bacterial proliferation causes competitive utilization of vitamin B12 (SCHJONSBY, 1989) and deconjugation of bile salts, and (4) pancreatic exocrine function is impaired with reduced secretion of lipase, trypsin, chymotrypsin, and amylase. Motility aberrations in PEM include delayed gastric emptying and prolonged intestinal transit time. Altered endogenous nutrient utilization is manifest in fatty infiltration of the liver.

Children with PEM are more susceptible to infectious diarrhea! diseases (GORDON, GUZMAN and ASOCOLI, 1964; TOMKINS, 1981; TROWBRIDGE, NEWTON, CAMPBELL, 1981; CHOWDHURY et al., 1990) which are a major cause of morbidity and mortality in developing countries. Diarrheal disease in children accounts for an estimated 4 to 4.5 million deaths per year (approximately 12000 deaths each day) in developing countries. Longitudinal studies indicate that children in developed countries sustain one to three episodes of diarrhea! disease per year, while children in developing countries have 3 to 10 episodes of diarrhea each year during their first 5 years of life, i.e., a total of 15-50 episodes averaging 4-6 days each (GLASS et al., 1991). From 3 to 20% of these acute diarrhea! incidents can become persistent (approximately 10% receive specific therapy), leading to a downward spiral of diarrhea, increasing nutritional deficits and hence greater susceptibility to diarrhea (CLAESON and MERSON, 1990).

Recurrent episodes of diarrhea lead to malnutrition as a result of anorexia with fever, catabolism from infection, malabsorption of nutrients and lack of adequate energy and protein intake because of the widespread custom of 'starving diarrhea'. It is estimated that 20-25% of the growth differential between children in developed and developing nations is due to acute, infectious diarrhea (MARTORELL et al., 1975; BLACK, BROWN and BECKER, 1984). Table 1 illustrates the potential magnitude of global food energy wastage due to acute, infectious diarrhea and associated clinical and subclinical malabsorption. This malabsorption is superimposed upon an average daily energy intake (Bangladesh Food Consumption Surveys) that is grossly inadequate (55 % of estimated energy needs of children of the same age).

Table 1. Calculus of potential food energy wastage associated with diarrhea! disease and subclinical malabsorption in children

Estimated energy intake in Bangladesh per day

67 kcal/kg

Estimated energy intake in Bangladesh per year

24455 kcal/year


35 days with diarrhea/year (7 episodes/5 days each)

70 days of recovery/year (10 days after each episode)

35 days X.67* X 67 kcal/kg = 1571 kcal/kg/absorbed during 35 days

70 days X.80* X 67 kcal/kg = 3752 kcal/kg/absorbed during 70 days

260 days X.94 X 67 kcal/kg = 16375 kcal/kg/absorbed during 260 days


21698 kcal/kg/absorbed during one year

* 67% of kcal absorbed during the diarrhea! episode. 80% during recovery (MOLLA et al., 1986).

# Without subclinical malabsorption approximately 6% (the average of 6.7 and 5.4) of the yearly food energy are malabsorbed, i.e., 94% are absorbed.

Note: The assumptions are consistent with published data from the International Center for Diarrheal Disease Bangladesh (ICDDB) for a large population in the Matlab District studied over many years (MOLLA et al., 1986), but the assumptions will obviously vary with time, place and population.

Subclinical malabsorption is more difficult to define and estimates of its current prevalence are elusive. Subclinical malabsorption is related to intestinal pathology or structural abnormality, termed tropical enteropathy, which is found on biopsy combined with altered functional tests of absorption (xylose, glucose, sucrose, nitrogen and fat). In the early 1970s almost every study involving a general population in a tropical area demonstrated a prevalence rate of tropical enteropathy between 30 and 50% (up to 80%), but most of these studies were based on abnormal xylose absorption tests (ROSENBERG and SCRIMSHAW, 1972). Studies in tropical and subtropical regions of Latin America, Africa, Asia and the Near East show a high percentage of asymptomatic, low-income people with non-specific intestinal abnormalities; increase in inflammatory cell infiltrate, moderate blunting and shortening of intestinal villi, and crypt elongation (SCHENK, KLIPSTEIN and TOMASINI, 1972; MAYORAL et al., 1972; SCHNEIDER and VITERI, 1972). Because in the extreme form these types of lesions are evident in pathologies associated with obvious malabsorption, it may be hypothesized that in an attenuated form they may produce subclinical malabsorption.

These morphological alterations in the intestine are acquired (LINDENBAUM, HARMON and GERSON, 1972; KEUSCH, PLAUT and TRONCALE, 1972) and appear to be due to repeated enteric infection. 130th malnourished and better nourished Bangladeshi children demonstrated hypochlorhydria, but the better nourished group did not have the gastric bacterial colonization evident in the malnourished group (GILMAN et al., 1988). Correcting nutritional deficits results in normalization of stomach and intestinal morphology and microflora in some circumstances, but this is not a universal observation in children and adults. Functional absorption tests in these areas show a high prevalence of xylose, lactose, and fat malabsorption; impaired absorption of nitrogen and vitamin B12 are less well documented.

It is difficult to interpret many of these functional absorption studies, since the children or adults are often not consuming their usual diets, and the food they eat is generally not measured accurately. Further, functional parameters are usually not quantified, only categorized as abnormal or normal. There does not seem to be any simple relationship between enteric infections, altered intestinal morphology or abnormal functional intestinal absorption tests (i.e., xylose) and reduced utilization of food nutrients.

One study (COWAN, 1972) did quantitate the level of subclinical malabsorption in 31 non-pregnant females and 29 males (well-nourished) aged 16 to 65 years in a Punjab village. Fifty-three percent had steatorrhea ranging from 6.1 to 13.4 g/d (mean excretion of 7.7 g/d). Seventy-two percent demonstrated xylose malabsorption. The combined nutrient losses amounted to about 150 kcal (628 kJ). This would represent 6.7% of the estimated energy requirement of a 35-year-old rural woman in a developing country, or 5.4% of the estimated energy requirement of a 25-year-old subsistence farmer, even without taking into consideration possible excess nitrogen losses (FAO/WHO/UNU, 1985).

Rather than focusing on altered intestinal morphology or abnormal functional tests of absorption, the relevant question becomes what percentage of the world's population is malabsorbing enough of their usual intake of nutrients to adversely affect their nutritional status? Even a subclinical malabsorption prevalence of 5% would have significant consequences on food energy and possibly protein wastage as well as growth potential and predisposition to infection. For example, xylose malabsorption was associated with growth retardation, increased incidence of severe weanling diarrhea and increased incidence of live-born sibling death in East Pakistani children (HARPER, 1972; EINSTEIN, MACKAY and ROSENBERG, 1972).

The consequences of either clinical or subclinical malabsorption undoubtedly vary with the adequacy of nutrient intake and presence or absence of concurrent infection. Pakistanis and Indians who were better nourished had less evidence (normal serum albumin, serum folate, hemoglobin and serum B12) of the impact of subclinical malabsorption than more poorly nourished Haitians (ROSENBERG and SCRIMSHAW, 1972).

In rural and underdeveloped areas where intake is inadequate or marginal and there is endemic intestinal infection, subclinical malabsorption may act as the final stress to tip the scales toward PEM. BRIEND et al. (1989) found 6 to 35-month-old Bangladeshi children were free of diarrhea 90.4% of the time, but even during diarrhea-free intervals, weight gain was only equivalent to 74.5% of the National Center for Health Statistics (NCHS) median. This points to some combination of inadequate intake, poor nutrient content and bio-availability, and subclinical malabsorption.

The Food and Agriculture Organization (FAO) and World Bank provide estimates of the global prevalence of PEM between 450 million and 1 billion persons. The Fifth World Food Survey (FAO, 1985) indicated high percentages of various populations who daily do not meet their physiologic energy needs (Table 2). These numbers may provide a more realistic estimate of those at nutritional risk because of marginal or inadequate intakes and subclinical malabsorption. Because of demographic and political changes, these numbers may be larger, but are unlikely to be smaller.

Table 2. Estimates of various populations at nutritional risk secondary to inadequate or marginal intake (FAO, 1985)




Far East

210 million



70 million


Latin America

38 million


Near East

16 million

2. Diarrheal diseases

Even under conditions of apparent adequate intake, infection can precipitate nutritional deficits because of increased basal metabolic rate (BMR) due to fever, cannibalization of skeletal muscle for amino acids secondary to impaired ketosis, and negative nitrogen balance even in asymptomatic infection (BEISEL, 1972; 1985). Infection can produce nutritional deficits without clinical diarrhea because its nutritional sequelae evolve from the production of cytokines (IL-1 and TNF) by stimulated immune system cells. For this reason, colonization with enteric pathogens that cause inflammatory diarrhea, combined with subclinical malabsorption and marginal or inadequate intake, may be the biggest contributor to the development of nutritional deficits and growth retardation. In this case, one would have marginal or inadequate exogenous nutrient intake, altered endogenous nutrient metabolism secondary to cytokine production and increased nutrient losses via malabsorption.

Different enteric pathogens may be associated with varying levels of disease severity and thus exacerbate the effects of marginal intake and subclinical malabsorption to greater or lesser degrees. It is known that specific pathogens do differentially affect malabsorption. For example, SACK et al., (1982) found carbohydrate malabsorption with rotavirus but not enterotoxigenic Escherichia coli diarrhea.

Enterotoxigenic Escherichia coli and rotaviruses are the most common diarrheal etiologic agents in developing countries. Rotavirus causes diarrhea with equal incidence among children in developed and developing countries, regardless of the level of hygiene or the quality of water or sanitation. Norwalk-like viruses, Campylobacter jejuni and Clostridium difficile are increasing in incidence in developed areas. Shigella, Salmonella, Cryptosporidium species and Giardia lamblia are found throughout the world. Clostridium difficile and Salmonella are the most common nosocomial diarrheal infections with Campylobacter jejuni, Shigella and Cryptosporidium being reported with increasing frequency in hospitals in both developed and developing areas (GUERRANT et al., 1990).

Special settings where diarrhea! diseases and concerns regarding malabsorption are becoming increasingly important are hospitals, chronic care facilities for the elderly, child care centers and in persons with acquired immunodeficiency syndrome (AIDS). Diarrhea is reported to occur in 30 to 50% of persons with AIDS in the United States and in up to 90% of cases in Africa and Haiti (SMITH and JANOFF, 1988). The World Health Organization (WHO) estimates that 8 to 10 million adults and 1 million children throughout the world are infected with the human immunodeficiency virus (HIV), the etiologic agent of AIDS. It is projected that by the year 2000, 40 million persons may be infected with HIV and more than 90% of these persons will reside in developing countries within sub-Saharan Africa, South and Southeast Asia, Latin America and the Caribbean (The HIV/AIDS Epidemic, 1991), where the risk of inadequate nutrient intake and subclinical malabsorption is greatest.

3. Nutritional recommendations

Since vaccines are not available for the two most common causes of dehydrating diarrhea in children under 5 years of age (rotavirus and enterotoxigenic E. coli), the most effective interventions to reduce death and morbidity remain rehydration and refeeding. The recommendations of discontinuing milk feeds and solid food during diarrhea was based initially on observations of increased volume and frequency of stools and transient lactose malabsorption during diarrhea. Research during the last decade has overturned these misconceptions by showing that rapid reintroduction of full strength cow's milk and weaning foods does not prolong the duration of acute diarrhea or appreciably increase stool volume (PLACZEK and WALKER SMITH, 1984; DUGDALE et al., 1982; HAQUE, AL-FRAYH and EL-RIFAI, 1983), yet it results in better weight gain. Clinically, children often present with acidic stools (pH < 5.0) and increased reducing substances (lactose), but this malabsorption appears to have clinical relevance in only a few (CLAESON and MERSON, 1990).

Studies in Peru showed that mild-to-moderately malnourished children who receive early feeding during acute diarrhea with a full strength diet (lactose-free synthetic liquid formula) had better net absorption of nitrogen, fat, carbohydrate and total energy; better retention of nitrogen; increased incremental body weight gain, arm circumference and skinfold thickness when compared to groups that received gradual refeeding or were starved for 24-48 hours upon admission (BROWN et al., 1988).

BECKER et al., (1991) found that dietary energy intake per kilogram body weight had a more significant impact upon weight gain in 5 to 18-month-old Bangladeshi children than either fever or diarrhea. Based on their data, these investigators projected that, if energy intakes were increased to WHO recommended level and diarrhea was kept at low levels (1 to 3 episodes per year), then weight gain would approach the NCHS reference population median. This implies that with the current diarrhea! prevalence in developing countries, one would have to exceed the WHO recommendations to increase weight. Since nutrient losses with subclinical malabsorption in some studies approximate those in acute, infectious diarrhea, exceeding the WHO recommendations for energy may be applicable here also. This study essentially used dietary energy as a proxy for total dietary intake and it is conceivable that the presence, amount and bioavailability of other nutrients are equally important.

LUTTER et al., (1989) found that protein-calorie supplementation from birth to 36 months completely offset the adverse effects of diarrhea on incremental linear growth. The protein-calorie supplement and diet provided 132% of the WHO energy recommendation and 459% of the WHO protein recommendation at 36 months. Children with very low levels of diarrhea did not benefit from supplementation in the same way as children with more severe diarrhea; the beneficial effects of supplementation appeared to increase as diarrhea worsened. These children still did not achieve the NCHS medians for weight or height, and the authors postulate one reason may be irreparable stunting from inadequate intake during the first 18 months, when dietary intake was unmeasured, combined with continued exposure to a high-risk environment (which may include subclinical malabsorption). In this study one cannot separate the effects of energy and protein since protein in excess of needs would either be oxidized for energy or shunted into fat deposition. Further, nitrogen retention improves as malnutrition worsens, which may have been partly responsible for the differential outcomes between the various diarrhea! severities.

These studies provide some frame of reference for the level of dietary energy and protein needed to overcome the deleterious effects of diarrhea! disease in young children. Such information is not available for adults or for subclinical malabsorption. Nutritional recommendations designed to overcome the adverse effects of either clinical or subclinical malabsorption must consider the following complexities: (1) the adequacy of exogenous nutrient intake, (2) mobilizable endogenous nutrient reserves (lean body mass and fat), (3) cytokine producing effect of enteric pathogens and effect on host nutrient metabolism, (4) malabsorption type and severity, and (5) known protein-energy interactions.

Protein-energy interactions include: (1) at any given energy intake, increasing the protein intake will improve nitrogen retention, (2) at any given protein intake, increasing energy will improve nitrogen retention, (3) the efficiency of nitrogen retention will parallel the degree of malnutrion, and (4) in the hypermetabolic phases of infection, because of increased catabolism, there will be less nitrogen retention for any given energy and protein intake. Further, BRUNSER et al., (1966) showed that, when protein intake is decreased and caloric intake is normal in malnourished infants, intestinal morphologic changes are more marked than when both protein and caloric intakes are decreased.

Outcomes in investigations designed to assess nutritional needs in clinical or subclinical malabsorption should be evaluated under usual living conditions and dietary intake. As an example, there is a high prevalence of lactose malabsorption in tropical and subtropical areas when assessed with lactose tolerance tests (administration of 55 g pure lactose). In contrast, there is apparent tolerance of milk when taken in small amounts (250-350 cc or 12-17 g lactose), when used in nutritional supplements or when taken with food.

Steatorrhea also seems to be highly prevalent in many of the studies evaluating subclinical malabsorption, but in most of these investigations subjects were given diets that were quite high in fat relative to normal consumption in these areas. No mention of type of fat or level of dietary fiber intake is made, both of which may affect fat absorption. Fat malabsorption would have greater impact upon energy nutriture than carbohydrate malabsorption because of the higher energy density of fat. There are also predictable vitamin and mineral losses accompanying steatorrhea: fat soluble vitamins (A, D, E, K); calcium, zinc, magnesium, and iron secondary to soap formation with divalent cations; and folic acid and/or vitamin B12 if there is ileal involvement.

Nutritional surveys in tropical and subtropical areas have demonstrated low serum retinol and zinc as well as iron deficiency and macrocytic anemia; inadequate intake and parasitic infection are characteristically blamed, but perhaps steatorrhea is contributory. Further, it is conceivable that nutrients from foods may be absorbed to differing degrees. KEUSCH (1972a; b) found significantly higher serum albumins in milk-drinking Thai orphanage children when compared to a group of comparable non-milk-drinking village children, although their degree of xylose and lactose malabsorption was the same. Absorption of glucose, sucrose and fat was normal in both groups. Protein, in this case, may have been absorbed well, even though functional absorption tests show carbohydrate absorption was impaired. In a recent study in Guatemala, rice cereal with milk was well absorbed by 7 of 8 infants with acute gastroenteritis, and such readily available foods are as effective in the nutritional management of acute diarrhea.

During febrile illness children develop anorexia, but appetite and intake remain good during diarrhea! illness and recovery, if sufficient foods of appropriate energy density are provided (Nutrition Reviews, 1991). The only certain nutritional recommendation appears to be that continued feeding during diarrhea! illness and aggressive feeding during convalescence or subclinical malabsorptive states is essential to optimize nutrient balance, promote intestinal mucosal renewal, ameliorate growth retardation and prevent recurrent infection.

Contents - Previous - Next