Encyclopedia of Food & Culture | Malnutrition: Protein-Energy Malnutrition

MALNUTRITION: PROTEIN-ENERGY MALNUTRITION. Protein-energy malnutrition (PEM) may be present at any time during the life cycle, but it is more common in the extreme ages, that is, during infancy/childhood and in the elderly. The present review will be restricted mostly to the condition present during infancy and childhood.

Protein-energy malnutrition is a syndrome characterized by its progressive onset and a series of symptoms and signs that encompass a continuum, ranging from clinically undetected manifestations to the full-blown clinical picture of marasmus or kwashiorkor. A syndrome is defined in clinical practice as a set of symptoms and signs that may be caused by different etiologies.

In the case of PEM, the earliest symptoms include subtle changes in the mood of the child, which may be described by the mother as saying that the child is not as playful as he/she used to be. Further changes include a loss of appetite and a loss of interest in the surroundings, which lead to decreased social interaction with peers or siblings and adults (parents or other caregivers). When PEM becomes more severe, there are adverse effects on the child's cognitive and behavioral development, evident in both the short and the long term.

In relation to signs, the earliest clinical sign of PEM is the lack of adequate weight gain. Also common in the early stages are mild episodes of common acute infectious diseases, such as acute diarrhea or acute respiratory infections. As the condition advances, the child will show signs of body wasting, progressing to an extreme thinness. If the syndrome becomes chronic, there are small or no increases in length. When the condition becomes more severe, the child may show the clinical pictures of marasmus or kwashiorkor, which will be defined later in this article.

The etiology of protein-energy malnutrition as a syndrome may be classified as primary or secondary. Although in practice most cases of PEM are caused by a combination of both, the concept may be useful for targeting interventions. Primary PEM refers to a deficit of available food. This, in turn, may be because of biological conditions, such as maternal malnutrition prior to or during pregnancy and lactation, or to social conditions, such as poverty; to a limited or selective unavailability of food; to war; to ecological disasters leading to famine, or, more often, as a result of profound social inequalities, either at the individual level (discrimination, refugees, prisoners) or at the community or country level. The largest prevalences of protein-energy malnutrition are found in socioeconomically deprived areas of the world, as will be reviewed further on in this article. Secondary causes of PEM include several conditions that impair food intake, absorption, or utilization, or that increase energy and/or protein requirements or losses. Secondary causes of PEM may be biological or social conditions.

Biological conditions may interfere with food intake, such as congenital anomalies (for example, harelip); with absorption, such as any of several malabsorption syndromes (for example, tropical sprue); or with utilization, such as inherited metabolic diseases (for example, phenylketonuria). Biological conditions that increase the need for energy include all infectious diseases accompanied by fever, and other diseases that increase catabolism, such as tuberculosis, or that are accompanied by an increased nutrient loss, such as intestinal parasitism.

On the other hand, social causes that affect food intake, whether it be in quantity or quality (protein-energy or micronutrient content), include several conditions associated with poverty, such as ignorance, inadequate weaning practices, child abuse, alcoholism or other drug addictions, and others.

Different conceptual frameworks for the study of malnutrition have been proposed and adopted throughout the years; one of the most widely accepted ones was developed during the WHO/UNICEF Joint Nutrition Programme in Iringe, Tanzania (B. Jonsson et al., 1993) (UNICEF 1990), from where it has been extended to many parts of the world. An appealing feature of this conceptual framework is that it may be adapted to describe causes of the major nutritional deficiencies present in the world, including vitamin A, iron, and iodine deficiency.

Clinical Picture of Marasmus and Kwashiorkor

Marasmus is characterized by a chronic and severe restriction of both energy and protein to the body. Marasmus is more frequently found at a younger age than kwashiorkor, usually in children under one year of age. A marasmic child presents severe wasting, with a very low weight-for-age and reduced length-for-age, often below -3 standard deviation of the reference population values. The clinical history of a marasmic child may reveal poverty or famine affecting the family; inadequate child-rearing practices, like starvation wrongly prescribed as part of the management of diarrhea; an early stopping of breast-feeding; over-dilution of formula; a history of repeated and/or chronic infections, such as diarrhea or tuberculosis; or some physical condition that affected the child's growth and development, such as prematurity, mental defects, or a malabsorption syndrome. The mother or caregiver will often report that the child is hungry. On a first appreciation, the child may be interested in the environment, with an active cry and reaching for food if offered, or else he/she may be depressed to the point of coma. The typical clinical picture is that of a striking loss of subcutaneous fat and muscle wasting, observed as markedly thin limbs, an evident rib cage, sunken cheeks and eyes that give the child a "monkey-like" or gaunt appearance, a prominent abdomen (although with no evidence of an enlarged liver), and a relatively big head. The hair is often thin and dry, and comes off easily. However, skin rashes or dermatosis are not usually present. Common micronutrient deficiencies include vitamins A and D, zinc, and iron, although anemia is less common in marasmus than in kwashiorkor.

The clinical picture of kwashiorkor is not as striking in appearance, as this syndrome often affects slightly older children, i.e., between one and three years of age. The clinical onset of kwashiorkor usually takes place in a shorter period of time as compared to marasmus, and is characterized by a relative, though severe, limitation in protein intake, with a lesser involvement of energy deficit. A child affected with kwashiorkor is often apathetic to external stimuli, is irritable, and gives the impression of misery, rejecting or crying when cared for. The most salient clinical characteristic of this syndrome is the presence of edema, which may mask the evidence of body weight loss and reduced length in relation to age. It is also common to find skin lesions that range from a flaky, pink dermatosis with skin dryness and depigmentation to deep ulcerations. Also common are petechiae and ecchymoses, as well as clinical signs of anemia. The hair presents discoloration with bands of dark and light hair (described by clinicians as the "flag sign"). An enlarged, fatty liver is also characteristic of kwashiorkor, palpable as a soft mass under the right rib cage. Co-occurring micronutrient deficiencies are common, so clinical signs of deficiencies of specific vitamins, including A, B, C, D, iron, or others, also may be present.

Classification of Protein-Energy Malnutrition

The clinical signs of severe PEM are so impressive that, for several years, they drew the attention of pediatricians and other physicians interested in furthering the understanding of the clinical syndromes and their treatment. Therefore, the study of PEM was long confined to the hospital setting. Actually, the first classification of PEM came from Mexican observers, who ranked the severity of malnutrition based on the risk of death for children with a clinical diagnosis of PEM. This group, led by Gomez, proposed that children with a weight-for-age deficit greater than 40 percent in relation to a reference population were in the greatest risk of dying, and thus labeled them as having third-degree malnutrition. Further, children with 25–40 percent weight-for-age deficit were labeled as having second degree malnutrition, and children with 10–25 percent weight-for-age deficit were classified as having first degree malnutrition (Gómez-Santos, 1946).

This classification had a high predictive value for the risk of death, and therefore had important implications for clinical practice. It was further abused, however, when its use was extended to the classification of malnutrition at a population level. In other words, children with no evidence of clinical malnutrition who have low weight-for-age should not be classified as malnourished; doing so may not only misdiagnose an individual, but may over-estimate the prevalence of malnutrition in a population. Also, the Gomez classification has been criticized because a single measure of a child's weight referred to age gives no idea about the nutritional history of the child. That is, an underweight child may be growing according to his/her normal growth channel, may be recovering from a recent episode of weight loss ("catch-up growth"), or may be deteriorating in relation to the recent past.

In order to overcome these caveats, Waterlow proposed combining weight-for-height, as an indicator of an acute episode of malnutrition, with height-for-age, as an indicator of chronic nutritional deficits that would be reflected in growth stunting (Waterlow, 1972).

Although these classifications have been used for several years, they have two important disadvantages that often are overlooked. To illustrate the first disadvantage, it is important to highlight the concept of Z-scores as a means of describing an individual child's anthropometric indicators in a normal distribution. The normal distribution of a reference population has been published by the World Health Organization (WHO) and is most often accepted worldwide as the standard for comparison. Eighty percent of the median weight-for-age might be above or below -2 Z-scores, depending on the child's age. The second disadvantage is that, to approximate a fixed point in the normal distribution, say, -2 Z-score, different percents of median have to be used depending on the anthropometric index used—for example, 90 percent for low height-for-age, or 80 percent of low weight-for-height.

In consequence, the World Health Organization Expert Committee on Physical Status has recommended the use of Z-scores to express weight-for-age, weight-for-height, or height-for-age relative to values reported in a reference population (WHO Expert Committee on Physical Status, 1995). The use of this system has several advantages; i.e., when applied at a population level, it allows the mean and standard deviation to be calculated for a group of Z-scores, and it allows the use of fixed cut-off points (i.e., -1, -2, or -3 Z-scores) to classify mild, moderate, or severe deficits for any anthropometric indicator. Although the use of Z-scores may be difficult to grasp for those who have been accustomed to classifying nutritional deficits based on the percent of median, the advantages of Z-scores outweigh their disadvantages.

Global Prevalence of Protein-Energy Malnutrition

The most recent estimates about the distribution of PEM at a worldwide level were compiled by the World Health Organization (WHO) Programme of Nutrition, available in its Global Database on Child Growth and Malnutrition (de Onis and Blössner, 1997). This database covered 95 percent of the total population of children under 5 years of age who lived in 103 developing nations in 1995, as was reported in nationally representative surveys available at the time. According to these data, an estimated 206.2 million children, who represent 38 percent of all children under 5 years old, were stunted (low height-forage); 167.3 million children (31 percent) were underweight (low weight-for-age), and 48.8 million children (9 percent) were wasted (low weight-for-height). PEM is most often found in the poor regions known as the "developing world." The largest number of affected children were found in Asia, where 41 percent of all under 5 years old were stunted, 35 percent were underweight, and 10.3 percent were wasted. Africa had 38.6 percent stunted, 28.4 percent underweight, and 8 percent wasted children of all those under 5 years old; Latin America and the Caribbean showed 17.9 percent stunted, 9.5 percent underweight, and 3 percent wasted children of all those under 5 years old. The proportion of children under 5 years of age affected in Oceania was 31.4 percent, 22.8 percent, and 5 percent, respectively, but the total number of children living in this region is much lower, so in reality, these percentages translate into many fewer children affected than in the other regions.

Since the mid-1980s, the Administrative Committee on Coordination/Sub-Committee on Nutrition (ACC/ SCN) of the United Nations periodically has examined the trends of malnutrition in the world's children. In its Third Report on the World Nutrition Situation (ACC/ SCN, 1997), this Committee (from data from 61 countries) estimated the trends in stunting with two or more nationally representative surveys. In the period from 1980 to 1995, stunting declined globally at a rate of 0.54 percentage points per year. Sub-Saharan Africa had an increase of 0.130 percentage points per year in the average prevalence of stunting; the remaining regions of the world showed statistically significant decreases that ranged from -0.26 in Middle-America and the Caribbean to -0.90 in Southeast Asia (Table 1).

The same Committee was able to use data from 95 countries that had data from at least one national survey to estimate the prevalence of undernutrition; underweight and stunting showed a consistent 11.5 percentage point difference. The higher prevalence was for the underweight classification. During the 1980–1995 period studied, only sub-Saharan Africa had an increase in the prevalences of both stunting and underweight; all the other regions showed decreasing trends in these two indicators (Table 1).

Acute and Long-Term Consequences of Protein-Energy Malnutrition

PEM results from a relative deficiency of protein (essential amino acids and/or total nitrogen) and energy substrates (carbohydrates, fats, or proteins). However, these deficiencies are almost always accompanied by micronutrient (minerals and vitamins) deficits. Manifestations of PEM differ depending on the duration, the severity, and the combination of these deficiencies. In the early stages,

there are functional impairments, which are later followed by biochemical and physical damage.

The identification, understanding, and treatment of the full-blown clinical syndromes characteristic of severe PEM began in the mid-1930s with the description of kwashiorkor (Williams, 1933). On the other hand, the identification and understanding of the functional manifestations of malnutrition have only come about during the last three decades of the twentieth century, with the launching of two large-scale, community-based research projects: the first one, known as the INCAP Longitudinal Study, was based in Guatemala (Habicht and Martorell, 1992). The second took place simultaneously in three countries—Egypt, Kenya, and Mexico—and was known as the CRSP study (Calloway, Murphy, et al., 1988).

Functional consequences of protein-energy malnutrition. As described earlier, the functional consequences of PEM were recognized and studied only relatively recently (Allen, 1993). Among the most well documented functional consequences of PEM are growth impairment, a reduced immune response, and a disruption in cognitive ability.

Growth impairment. Growth failure because of PEM usually starts to manifest very early in life. Information from the INCAP longitudinal study, as well as from the CRSP studies, coincides in showing that growth stunting begins at about 3 to 4 months of age and is complete before 18 months (Allen, 1995). A further contribution from the INCAP study was provided by a long-term follow-up of the same populations that showed not only that growth stunting present during infancy carried on until adolescence, but also that length at 3 years of age was a strong predictor of adolescent size (Martorell, Schroeder, et al., 1995). It also seems as if stunting in early life is correlated significantly with reduced physical performance (Haas, Martinez, et al., 1995) and reduced psychomotor and mental performance, both during late childhood (Mendez and Adair, 1999) and even until adolescence (Grantham-McGregor, 1995; Pollit, Gorman, et al., 1995).

Two more relevant issues related to growth failure are that there is a window of opportunity for intervention from the ages of 3 to 6 months, when response to the intervention may be greatest (Lutter, Mora, et al., 1990), and that most of the growth deficit found at later ages accumulated during the first months of life (Rivera, Cortes, et al., 1998).

Immune response. It has been recognized that malnutrition is the most common cause of immunodeficiency worldwide (Chandra, 1991). Actually, malnutrition and infection interact in a vicious cycle: the presence of one more easily leads to the development of the other (Scrimshaw, Taylor, et al., 1968). There are several mechanisms involved in this relationship. PEM impairs cell-mediated immunity, phagocitic function, and the complement system. It also diminishes immunoglobulin (IgA, IgM, and IgG) concentrations, and cytokine production (Chandra, 1991). Micronutrient deficiencies associated with PEM also adversely effect the immune response. For example, iron plays an important role in several metabolic functions, including both the host and invasive bacteria. Several microorganisms that infect the human body only achieve their full infectious activity in the presence of iron. Such is the case of bacteria that cause diarrheal disease, such as Escherichia coli, Yersinia septica, Salmonella sp., and Vibrio cholerae; and others responsible for lower respiratory infections, such as Mycobacterium tuberculosis, Klebsiella pneumoniae, Pseudomona aeruginosa, and Listeria monocytogenes. These microorganisms actively seek iron in their host during infection, uptaking it from destroyed red cells (erythrocytes) and body stores (liver). On the other hand, the host tries to make iron less available to invasive microorganisms, sequestering it through different mechanisms—referred to as nutritional immunity—that include the binding of iron to transferrin and lactoferrin, and the increase in ferritin saturation in the liver (Kochan, 1976). Other micronutrients that play active roles in modulating immunity include zinc, selenium, copper, vitamins A, C, E, B _6, and folic acid (Nezu and Nakahara, 1994).

Conversely, infectious diseases lead to malnutrition by several mechanisms that often interact with each other. Almost every malnourished child will sooner or later present with diarrhea. Many of the interactions between malnutrition and infection are understood because of studies of diarrheal disease (Chen, 1983); hence the illustration of the mechanisms by which these two morbid conditions interact is particularly useful. One of the first symptoms of diarrheal disease is anorexia, as a result of vomiting and abdominal discomfort. Also, fever, dehydration, and electrolyte imbalances contribute to it (Martorell, Yarbrough, et al., 1980). Anorexia leads to a restricted intake, which is often reinforced by erroneous caregiver practices. In part, it is culturally engrained in different societies to withhold food from a diarrheaaffected child (Bentley, 1988), and it is also quite common to find physicians who still think that it is necessary to "put the bowel to rest" during the acute stage of the illness (Brown and MacLean, 1984). The deleterious effect of decreased food intake is worsened by increased catabolic losses of nitrogen that occur as a result of increased metabolic rates and structural damage to the intestine (Powanda, 1977). Another consequence of intestinal damage is the transient loss of absorptive surface and absorptive function as a result of villous atrophy (Davidson and Barnes, 1979). This condition leads to a decreased absorption of macronutrients (fat and carbohydrates) and micronutrients (particularly fat-soluble vitamins). The presence of unabsorbed carbohydrates in the intestinal lumen increases the osmolarity of the intestinal content, thus causing an hyperosmolar diarrhea (Wapnir, 1982). It also subjects these substrates to bacterial fermentation, which produces gas and intestinal bloating, worsening gastrointestinal symptoms. Catabolic losses also are increased by the presence of fever. In cases of parasitic infestations, the child's nutritional status is impaired by blood losses that are secondary to colitis or direct intestinal mucosal damage (common in cases of roundworm [Ascaris lulmbricoides], hookworm [Ancylostoma duodenale and Necator americanus], or whipworm [Trichuris trichiura]). Parasitic infestations also are associated with respiratory symptoms (particularly in case of Ascaris infestations) and anorexia (Lunn and Northrop-Clewes, 1993).

Disrupted cognition. PEM can disrupt cognition in several ways. Following the lessons learned from the effect of PEM on the body during infections, the classic explanation was that malnutrition caused physical damage to the brain, particularly during sensitive periods of development, namely, during the first two years of life, when about 80 percent of the brain's growth is achieved (Guilarte, 1993; Levitsky and Strupp, 1995). At present, however, it is clear that there are several other mechanisms, aside from organic damage, by which malnutrition can impair intellectual development. There is also evidence that at least part of this damage may be reversible, even in the presence of structural damage to the brain (Levitsky and Strupp, 1995).

Malnutrition may affect brain growth and development, which will be reflected in cognitive disabilities, motor impairment, or lower intelligent quotient (IQ), by means of micronutrient deficiencies such as vitamin B ₆ or iron, both of which are vital for normal brain function (Guilarte, 1993; Pollitt, 1997). Malnutrition also may affect these functions because of energy deficiency, which limits activity and social interaction with peers and caregivers. This mechanism was explored first in the early 1970s by Levitsky and coworkers in a rat model. They showed that energy-deprived rats scored lower on such tests as maze running—a proxy for mental ability—because they were so feeble that they withdrew from contact with their peers and the objects in their surroundings (Levitsky and Strupp, 1995). Similar findings were shown to be present in children living in deprived third-world communities (Chávez and Martínez, 1982).

The extent to which PEM affects intellectual potential has been explored by studying the effect of protein-energy supplementation on behavioral development. In spite of different study designs that focused on prenatal supplementation (Rush, Stein, et al., 1980), on postnatal supplementation (Grantham-McGregor, Meeks Gardner, et al., 1990; Husaini, Karyadi, et al., 1991), or in both (Waber, Vuori-Chirstiansen, et al., 1981; Chávez and Martínez, 1982), results from these studies are consistent in showing that a significant proportion of the variability in mental and motor developmental scales during the first two years of life may be accounted for by nutritional supplementation.

The extent to which the differences in intellectual performance found at early ages in children affected by PEM carries on to later stages in life has been addressed by Pollitt et al. in a long-term follow-up study of Guatemalan children, who received supplements during the prenatal period and the first 2 years of life and were later followed up between the ages of 13 and 19 years old (Pollitt, Gorman, et al., 1995). This study found that children who had received a protein-energy supplement had significantly higher scores on tests of knowledge, numeracy, reading, and vocabulary, as well as a faster reaction time in information-processing tasks compared with children who had received only an energy supplement. This effect was particularly strong for protein-energy supplemented children at the lowest end of the socioeconomic distribution, an interesting finding when compared to only energy supplemented children, in whom the higher cognition test scores varied as a positive function of socioeconomic status, as expected. The authors interpretation is that the protein-energy supplement acted as a social equalizer in relation to the differences in performance usually found in populations as a function of differences in socioeconomic status.

Another long-term supplementation study was carried out in a Mexican village, where women received nutritional supplements during pregnancy and their offspring continued to receive micronutrient supplements from 12 weeks until 10 years of age. Compared to a control group (mothers and children from the same village, recruited two years before supplementation began), children who received supplements showed significantly better IQs, school performance, and behavior (Chávez, Martínez, et al., 1995).

The studies of the effects of iron deficiency on intellectual and motor abilities were addressed specifically during the 1980s and 1990s. Several well-designed intervention-control studies have shown that, before treatment, average mental scores on the Bayley Scales of Infant Development of infants with anemia were 6 to 14 points lower than the scores of non-anemic controls (Lozoff, Brittenham, et al., 1982; Grindulis, Scott, et al., 1986; Lozoff, Brittenham, et al., 1988; Walter, De Andraca, et al., 1989), and average motor development scores were 9 to 11 points lower, differences of statistical and clinical significance. No significant improvement on the test scores of initially iron-deficient children were noted following iron supplements for two to three months (Aukett, Parks, et al., 1986; Lozoff, Brittenham, et al., 1988; Walter, De Andraca, et al., 1989). Fewer studies have addressed whether these deficits prevail in later ages. In a long-term follow-up study of Costa Rican children at age 5 years whose iron status had been documented and consequently treated in infancy under careful supervision, Lozoff et al. found that at five years of age, all children had excellent iron status. However, those children who had been severely iron deficient during infancy (hemoglobin ≥100 g per liter) showed lower mental and motor functioning scores at school entry than did the rest of the children, even after controlling for background factors that were potential confounders (Lozoff, Jimenez, et al., 1991). Further, even anemic children with hemoglobin levels < 100 g per liter before and after treatment also had poorer outcomes at five years of age, compared to non-anemic children. Strong as this evidence may be, it is relevant to point out that, to date, there is no definite proof that iron deficiency is the cause of children's lower test scores. For obvious ethical reasons, the gold standard of experimental designs, the double-blind placebo-control study, has not been carried out.

Protein-energy malnutrition also may affect children's performance on cognitive tests by other, indirect mechanisms (often conceptualized as confounding variables in studies that attempt to establish links between PEM and impaired cognition).These include social and economical disadvantages (Johnston, Low, et al., 1987), differences in parental education (LeVine, LeVine, et al., 1991), years of schooling (Ceci, 1991), inadequate attention or affection from caregivers (Engle and Ricciuti, 1995), and other environmental factors, which may include peer interaction, parental presence in the home, etc. (Engle and Lhotska, 1999).

Recent research has addressed the role of breastfeeding (the gold-standard of good nutrition during the first months of life) on cognitive development, adjusting for the aforementioned variables. The results of a meta-analysis that included 11 studies that controlled for ≥ 5 covariates on the effect of breast-feeding on cognitive function, a statistically significant increment in cognitive function of 3.16 points was seen in breast-fed infants, consistent through all the studies, at 6 to 23 months of age. This study found a greater benefit of breast-feeding for cognitive development of premature babies (an adjusted benefit of 5.18 points), and a larger benefit in relation to duration of breast-feeding (an increase of the weighted mean benefit of 1.68 points with 8–11 weeks of breast-feeding to 2.91 points with ≥28 weeks) (Anderson, Johnstone, et al., 1999).

Concluding Remarks

Over the years, much has been learned about protein-energy malnutrition, its causes, and its effects. Without pretending that all is known, available knowledge can alleviate this burden on human development and social inequalities. Although the treatment of malnourished children all over the world is a clear imperative, the key to solving the problem is to focus on prevention. Preventive actions should be interdisciplinary. These actions should encompass a broad focus on education, particularly directed to women; they should include actions to improve sanitary conditions, schooling opportunities, employment, agricultural produce, and access to diverse food sources, particularly those rich in micronutrients. All sectors of society, including government and nongovernment organizations, should work together toward a common end. The opportunities to make a substantial improvement in the nutritional status of children all over the world are here, as never before in history. We have studied the causes of malnutrition, its mechanisms, and its consequences. It is now time to study the impact of specific interventions tailored to solve persistent problems.

Homero Martínez