Special Report
Starvation in the Modern World
N Engl J Med 1993; 328:1058-1061April 8, 1993
- Article
During his first visit to an underdeveloped country, a prominent psychologist spent hours quietly observing, and being observed by, a skeletal infant who at eight months of age weighed less than when she was born. Her slow, shallow breathing was almost imperceptible, her pulse rate very slow, and her blood pressure scarcely measurable. She seemed very near death. There were no discernible movements except those of her eyes, which warily tracked anyone entering the room. Irregularly breast-fed for less than a month and weaned to teas and broths, she had survived almost total starvation by a miracle of adaptation: stopping all growth, reducing basal oxygen consumption, severely depleting her cell mass, and eliminating all physical activity1. Despite her mistrust, which took weeks to overcome, she managed to drink progressively larger amounts of milk-based formula and was discharged from the hospital in good shape, not the usual outcome for most young victims of such severe starvation.
The manifestations of starvation depend on the context, the person's previous nutritional status, his or her age, and the severity of food deprivation. Morbidly obese adults who submit voluntarily to prolonged fasts consume only water, vitamins, and potassium salts for many months. After a few days ketosis stops their hunger, thirst abates, polyuria and then oliguria develop in some, and fecal volume decreases. Despite mild initial headache, nausea, and tension, they feel remarkably well. Physical energy decreases gradually; after some weeks, orthostatic hypotension with tachycardia becomes a problem. The levels of serum electrolytes, lipids, proteins, and amino acids remain unchanged because the subjects subsist on the ample stores of energy in their own bodies, balanced by amino acids derived from catabolism of their own proteins. Part of their lean body mass is sacrificed to the initial gluconeogenesis and to continued obligatory losses of nitrogen2.
When healthy, nonobese, young adult male volunteers had their energy intakes reduced well below their maintenance needs, to less than 1600 balanced kilocalories per day for six months, and lost 20 to 25 percent of their body weight, they became emaciated, lost much of their hair, and developed roughness and hyperpigmentation of the skin. Many had edema of the legs and genitalia, and less often of the face. They had marked decreases in pulse rate and blood pressure, and they became intolerant of cold; fatigue, muscle cramps, and soreness were common. They did not have diarrhea or an increase in respiratory infections3. In marked contrast to totally starved obese subjects, they had an increasing fixation on food and suffered from severe hunger pangs that distracted their attention from manual or intellectual pursuits. They lost the ability to concentrate and became moody, depressed, and isolated. In tests of intelligence, their performance was not affected. It took months of overeating for the almost insatiable craving for food and the altered behavior to disappear4.
The previously well-nourished victims of the famine in the Netherlands during World War II suffered a year of reduced food intake followed by six months of much more severe deprivation, compounded by the stresses of war and the German occupation. Those whose weight loss reached 40 percent almost always died. Infection played a minor part. Women in the first trimester of pregnancy aborted, and those in the third trimester gave birth to babies that were underweight for their length5. Listlessness and indecision characterized the survivors6.
The different syndromes seen in severely malnourished infants and young children exemplify the pathogenesis of starvation. An infant weaned very early in life to a grossly inadequate set of liquids soon burns up any stores of glycogen, suppresses the activities of insulin7 and somatomedin,8 mobilizes fat depots with the help of growth hormone, stops growing, limits energy expenditure by curtailing physical activity and thermogenesis, and begins to consume muscle and organ proteins. Of particular importance is the cessation of cell mitoses and migration in the intestinal mucosa, resulting in the presence of many fewer but still healthy-looking cells9. In muscle and most organs, the protein and RNA-to-DNA ratios are markedly reduced, as are the concentrations of intracellular cations, notably potassium, magnesium, and zinc10. These changes, typical of infantile marasmus (the lack of both calories and protein), also characterize total or nearly total starvation in children and adults.
When the principal cause of malnutrition is a dietary protein deficiency with a still substantial intake of energy, usually as the result of inappropriate weaning diets, other manifestations are evident: fatty liver, marked hypoalbuminemia, hyponatremia, edema, and a “flaky-paint” dermatosis1. In children this is known as kwashiorkor. It can be superimposed on marasmus when there is infection, notably diarrhea or measles. Instead of becoming dormant, many tissues, notably hair roots and intestinal mucosa,9 continue to multiply and produce grossly defective cells. Scarce amino acids are diverted from the liver and the proteins it synthesizes, such as albumin, in the vain attempt to continue growth. A dystrophic intestinal mucosa accounts for the markedly increased severity of lactose malabsorption (and hence, intolerance to milk), over and above that expected for age or ethnic origin or from marasmus alone. During the famine of 1968 in Biafra, the Ibos were cut off from fish and legumes, their major sources of protein, and for months subsisted on little more than their main staple: cassava, an almost protein-free root. A huge epidemic of kwashiorkor ravaged their children and made their rehabilitation with milk particularly hazardous. When the supplies of cassava were exhausted, the surviving children became marasmic11.
The “hunger edema” of adults and children in the Irish and other famines, and of the volunteers in the Minnesota study,3 had a complex cause. Although edema was commonly attributed to hypoalbuminemia alone, Hansen and Brock demonstrated that the provision of potassium without any protein produced effective diuresis and reduced edema in fully half their study subjects12. When starved people who have reduced concentrations of intracellular cations, particularly potassium, and relatively expanded extracellular compartments are rehabilitated with even small amounts of carbohydrate and substantial amounts of salt, gross edema and often fatal congestive heart failure can be precipitated13. The deaths of many survivors of Nazi concentration camps who were given free access to a variety of foods taught us caution in rehabilitation14. If high-carbohydrate feedings early in rehabilitation are infrequent, they can also lead to dangerous reactive hypoglycemia in people with marasmus who are hypersensitive to insulin; such feedings can also contribute to the fatty liver found in the “nutritional recovery syndrome.” In infants with marasmus, early oral feeding can cause repeated episodes of life-threatening necrotizing enterocolitis, which is precipitated by gas formed from incompletely absorbed and anaerobically hydrolyzed carbohydrates. Although kwashiorkor is life-threatening, it responds promptly to dietary treatment if there are no complications. When kwashiorkor is complicated by infection or marasmus, dietary treatment takes as long as treatment of severe marasmus alone.
Clinically apparent vitamin or mineral deficiencies are not common during famines, with the notable exception of scurvy in Ireland, Ethiopia, Sudan, and Somalia and of ophthalmia, due to vitamin A deficiency, in some Asian and African famines. Depletion of intracellular cations has been mentioned; the same is true for other essential minerals, such as iron and copper, particularly when there is chronic or repeated diarrhea15. In the case of iron, no more is accumulated or retained than is necessary to carry the oxygen needed by the much-reduced metabolically active cell mass16. The same can be said for all known essential minerals and vitamins: the amounts available at any time are seldom more than those needed to serve a severely shrunken organism.
When demand is increased against this background, as during an infection, the degree of depletion of vitamins and minerals becomes evident, and clinical deficiency syndromes appear. Injudicious attempts to rehabilitate the starved can result in catastrophe, for similar reasons17. High protein intake creates an increased requirement for potassium, magnesium, zinc, vitamin A, and water-soluble B-complex vitamins: pyridoxine, riboflavin, and niacin. Niacin deficiency has precipitated pellagra, as was seen recently in Mozambique. So great is the risk of inducing severe clinical deficiency of vitamin A by stimulating a high protein intake that a leading voluntary agency warns in its manual for field workers that dry skimmed milk, one of the most important resources, is likely to be devoid of vitamins A and D unless it comes from UNICEF, Canada, or the United States18. High calorie intake creates an increased need for intracellular cations, thiamine, and other water-soluble B vitamins. In the course of my work in Peru, I have seen clinical deficiencies of potassium, magnesium, zinc, iron, copper, vitamin A, ascorbic acid, niacin, pyridoxine, folic acid, and vitamin B12 brought on by diets that were inadequate in one of these nutrients, given the increased protein being consumed or the rate of weight gain.
For many years, the received wisdom has been that starving people should be provided with a supply of their customary staple food, generally a cereal. This method is not always possible or wise, however. Teply, of UNICEF, warned of the futility of this practice when, because of displacement or the destruction of their homes, the recipients could not process or prepare the staple in the established manner19. This is often true of the sorghum distributed in Ethiopia, Sudan, and Somalia, because fermentation or industrial extrusion is necessary to make it fit for direct human consumption20. Equally important is the availability of the usual complementary sources of protein, missing essential amino acids, vitamins, and minerals. In the Andean highlands, for instance, if the white maize or potato that serves as the staple is not accompanied by turnip greens, vitamin A deficiency soon becomes evident.
Although anecdotal evidence suggests that some would rather starve than eat an unfamiliar staple, evidence indicates that during a famine people will eat highly unusual foods. It is said that during a famine, rice-eating Bengalis rejected the millet that was made available; it is much more likely that they could not process or digest what to them was a strange, very hard, and tiny cereal grain. During the famine induced by the potato blight in Ireland in the 1840s, when maize was imported from the United States to replace the failed potato, it was of a flinty variety that required double milling in a country where mills were few and far between. Yet until mills were supplied, the starving Irish managed to pound, soak, and cook the maize at home until they could consume and digest it. The maize was deficient in lysine and tryptophan, but with milling and the usual complements available, it did a creditable job of sustaining adults and older children, if not the very young. A sharp increase in starvation and death supervened, however, when with the best of intentions and under the guidance of a famous French chef from London, the maize was replaced with a delicious soup served in soup kitchens that was very low in calories, nearly devoid of protein, and almost certainly loaded with salt21.
Sometimes the grain is provided in a form that the recipients cannot use effectively. For example, bulgur wheat, though popular in certain areas of the world, may be quite foreign in others. When not prepared properly, it may act like little stones that are passed intact in the feces. White wheat flour is of little use when there are no facilities for baking bread or making pasta.
In the early 1960s, when UNICEF and other agencies became increasingly involved in providing food assistance to developing countries, the consensus among their nutrition advisers was that most grains, and certainly all roots and tubers except the potato, could not by themselves satisfy the protein requirements of adults, and less still those of infants and children. To replace the legumes, fish, or cheese traditionally used to complement the staple, it became commonplace to combine the donated grain with soy or peanut flour, rich sources of much-needed fat and lysine-rich protein. To this day such combinations, some with added dry skimmed milk, and mixed with vegetable oils before consumption, are very important elements of relief programs. The most commonly used are the mixture of corn, soy, and milk with vitamins and minerals that has been made available by the United States and the Bal'ahar mixture of whole wheat, peanut flour, and chickpea flour with vitamins and minerals that has been developed in India22. These mixtures, like dry skimmed milk, have very high protein contents (over 20 percent of calories) and solute loads, and without the addition of vegetable oils they would be dangerous for infants and small children, even those who are well nourished.
It has been recognized for many years that the very young and the very old are the first victims of famine and that the rehabilitation of infants and young children cannot be carried out effectively with the grains or cereal-legume combinations that do an adequate job in older children and adults. For many years, surplus dry skimmed milk has been the standby of emergency feeding for young children. It has also been recognized that this milk does not satisfy essential requirements for fatty acids, has a low energy density, and presents an excessive load of solutes to the kidneys. In many cases its high lactose content produces dehydrating diarrhea. In the famine in Biafra, such was notoriously the case among the very large number of children with kwashiorkor11. For their initial therapy, “K-mix I,” a mixture of casein, sucrose, and vegetable oils, with added vitamins and minerals, was used successfully. K-mix II, used after the initial phase, contained a substantial proportion of the more readily available, cheaper dry skimmed milk11.
The corn-soy-milk mixture was found to be nutritionally adequate as the main source of protein and micronutrients for infants well along in convalescence, but not during their initial care, mainly because of inferior digestibility. The very high protein content of the mixture made its dilution with sugar and oil mandatory23. An instant and sweetened version that requires no cooking was found to be far superior for very young or malnourished children24. The original version and its variants are still the mainstays of emergency feeding for infants and young children, although the instant version is recognized as being much superior18.
In 1974 McLaren denounced the exaggerated emphasis on protein in treating malnutrition in infants and children25. Unfortunately, there was a reaction, which led to the acceptance of the view that the staple foods of the world, if consumed in amounts sufficient to satisfy the energy needs of infants and young children, would automatically satisfy their protein needs. Fortunately, in actual practice the relief agencies seem to ignore these concepts and generally use mixtures containing adequate amounts of protein. Sometimes, as when they recommend the addition of dry skimmed milk to sweetened condensed milk, they are too cautious; sweetened condensed milk has a long history of success in primitive circumstances. Its high sugar content protects it against bacterial contamination but still leaves it with a more than adequate content of high-quality protein.
It is unfortunate, however, that 25 years after the famines in Bihar and Biafra, when so much was learned, we still use exactly the same foods for rehabilitation. Some of the most appropriate are not even available to relief workers. Although we will spend upward of a billion dollars for the military to protect the relief supplies for the starving Somalians, we spend next to nothing in developing and evaluating simple foods that might increase the likelihood of survival of the starving. The wave of enthusiasm generated in 1984 for aid to the starving Ethiopians seems to have ended in self-congratulation and obliviousness. Let us hope that the same does not happen again after our relief efforts in Somalia. Agricultural research and advances in food technology and nutrition science are necessary to make safe, inexpensive food universally available.
Source Information
From the Instituto de Investigacion Nutricional, Lima, Peru, and the School of Hygiene and Public Health, Johns Hopkins University, Baltimore.
Address reprint requests to Dr. Graham at P.O. Box 205, Gibson Island, MD 21056.
- References
References
1
Viteri FE, Pineda O. Effects on body composition and body function, psychological effects. In: Blix G, Hofvander Y, Vahlquist B, eds. Famine, a symposium dealing with nutrition and relief operations in times of disaster. Uppsala, Sweden: Almqvist & Wiksells, 1971:25-40.
2
Drenick EJ ,Swendseid ME ,Blahd WH ,Tuttle SG . Prolonged starvation as treatment for severe obesity. JAMA 1964;187:100-105
Web of Science | Medline3
Keys A, Brozek J, Henschel A, Mickelsen O, Taylor HL, eds. The biology of human starvation. Vol. 1. Minneapolis: University of Minnesota Press, 1950.
4
Keys A, Brozek J, Henschel A, Mickelsen O, Taylor HL, eds. The biology of human starvation. Vol. 2. Minneapolis: University of Minnesota Press, 1950:769.
5
Stein Z, Susser M, Saenger G, Marolla F. Famine and human development: the Dutch hunger winter of 1944-45. London: Oxford University Press, 1975.
6
Scrimshaw NS . The phenomenon of famine. Annu Rev Nutr 1987;7:1-21
CrossRef | Web of Science | Medline7
Milner RDG . Insulin secretion in human protein-calorie deficiency. Proc Nutr Soc 1972;31:219-223
CrossRef | Web of Science | Medline8
Hintz RL ,Suskind R ,Amatayakul K ,Thanangkul O ,Olson R . Plasma somatomedin and growth hormone values in children with protein-calorie malnutrition. J Pediatr 1978;92:153-156
CrossRef | Web of Science | Medline9
Brunser O ,Reid A ,Monckeberg F ,Maccioni A ,Contreras I . Jejunal biopsies in infant malnutrition: with special reference to mitotic index. Pediatrics 1966;38:605-612
Web of Science | Medline10
Cheek DB ,Hill DE ,Cordano A ,Graham GG . Malnutrition in infancy: changes in muscle and adipose tissue before and after rehabilitation. Pediatr Res 1970;4:135-144
Web of Science | Medline11
Ifekwunigwe AE . Emergency treatment of large numbers of children with severe protein-calorie malnutrition. Am J Clin Nutr 1975;28:79-83
Web of Science | Medline12
Hansen JDL ,Brock JF . Potassium deficiency in the pathogenesis of nutritional oedema in infants. Lancet 1954;2:477-477
CrossRef | Web of Science13
Cahill GF Jr . Famine symposium -- physiology of acute starvation in man. Ecol Food Nutr 1978;6:221-230
CrossRef | Web of Science14
Mayer J. What kind of organization and what types of trained personnel are needed in the field? In: Blix G, Hofvander Y, Vahlquist B, eds. Famine, a symposium dealing with nutrition and relief operations in times of disaster. Uppsala, Sweden: Almqvist & Wiksells, 1971:178-85.
15
Graham GG ,Cordano A . Copper depletion and deficiency in the malnourished infant. Johns Hopkins Med J 1969;124:139-150
Medline16
Viteri FE ,Alvarado J ,Luthringer DG ,Wood RP II . Hematological changes in protein calorie malnutrition. Vitam Horm 1968;26:573-615
CrossRef | Medline17
Toole MJ ,Waldman RJ . Prevention of excess mortality in refugee and displaced populations in developing countries. JAMA 1990;263:3296-3302
CrossRef | Web of Science | Medline18
Appleton J, Save the Children Ethiopia Team, eds. Drought relief in Ethiopia: planning and management of feeding programmes: a practical guide. London: Save the Children, 1987:48.
19
Teply LJ. What kinds of foods should primarily be provided for young children, in famine? In: Blix G, Hofvander Y, Vahlquist B, eds. Famine, a symposium dealing with nutrition and relief operations in times of disaster. Uppsala, Sweden: Almqvist & Wiksells, 1971:165-9.
20
Graham GG ,MacLean WC Jr ,Morales E , et al. Digestibility and utilization of protein and energy from Nasha, a traditional Sudanese fermented sorghum weaning food. J Nutr 1986;116:978-984
Web of Science | Medline21
Famine in Ireland. In: Aykroyd WR. The conquest of famine. London: Chatto & Windus, 1974:34-6.
22
Ramalingaswami V ,Deo MG ,Guleria JS , et al. Studies of the Bihar famine of 1966-67. In: Blix G, Hofvander Y, Vahlquist B, eds. Famine, a symposium dealing with nutrition and relief operations in times of disaster. Uppsala, Sweden: Almqvist & Wiksells,1971;23
Graham GG ,Morales E ,Acevedo G ,Placko RP ,Cordano A . Dietary protein quality in infants and children. IV. A corn-soy-milk blend. Am J Clin Nutr 1971;24:416-422
Web of Science | Medline24
Graham GG ,Baertl JM ,Placko RP ,Morales E . Dietary protein quality in infants and children. IX. Instant sweetened corn-soy-milk blend. Am J Clin Nutr 1973;26:491-496
Web of Science | Medline25
McLaren DS . The great protein fiasco. Lancet 1974;2:93-96
CrossRef | Web of Science | Medline
- Citing Articles (8)
Citing Articles
1
H. Ashrafian, T. Athanasiou, J. V. Li, M. Bueter, K. Ahmed, K. Nagpal, E. Holmes, A. Darzi, S. R. Bloom. (2011) Diabetes resolution and hyperinsulinaemia after metabolic Roux-en-Y gastric bypass. Obesity Reviews 12:5, e257-e272
CrossRef2
Oliver E. Owen, Karl J. Smalley, Robert L. Jungas. 2011. Starvation. .
CrossRef3
Joanna Dowman, Ramakrishnan Arulraj, Ian Chesner. (2010) Recurrent acute hepatic dysfunction in severe anorexia nervosa. International Journal of Eating Disorders 43:8, 770-772
CrossRef4
Naomi Fukagawa. 2007. Protein-Energy Malnutrition. , 669-682.
CrossRef5
F. Valdés. (2006) Vitamina C. Actas Dermo-Sifiliográficas 97:9, 557-568
CrossRef6
Carmela De Caprio, Alessia Alfano, Ignazio Senatore, Luigi Zarrella, Fabrizio Pasanisi, Franco Contaldo. (2006) Severe acute liver damage in anorexia nervosa: two case reports. Nutrition 22:5, 572-575
CrossRef7
Joseph Krenitsky. (1996) Nutrition and the Immune System. AACN Clinical Issues: Advanced Practice in Acute and Critical Care 7:3, 359-369
CrossRef8
Kathie M. Dalessandri, Kathryn Boor. (1994) World nutrition—the great breadfruit source. Ecology of Food and Nutrition 33:1-2, 131-134
CrossRef
