Measuring nutritional status in relation to mortality

Public Health Classics

This section looks back to some ground-breaking contributions to public health, reproducing them in their originalform and adding a commentary on their significance from a modern-day perspective. To complement this monthstheme of the Bulletin, Mercedes de Onis reviews the 1956 paper by F. Gomez et al. on mortality in second and thirddegree malnutrition. The original paper is reproduced by permission of Oxford University Press.

Measuring nutritional status in relationto mortalityMercedes de Onis1

In 1956, Federico Gomez and colleagues describedthe clinical picture preceding death and the apparentcause of death in malnourished children admitted tothe Nutrition Department of the childrens hospitalinMexico City (1). Themain purpose of their article, aclassic in the history of nutritional sciences, was toprovide information on clinical profiles of childmalnutrition and their associated risk of mortality.What made this paper a landmark contribution wasthe use of a simple anthropometric measurement weight to develop an indicator (weight-for-age)and, on this basis, a classification of varying degreesof malnutrition. To do this, Gomez and hiscolleagues relied on the average theoretical weightthey had found amongMexican children (2). Patientswere classified into three groups according to severityof malnutrition, namely, first degree (7690% of thetheoretical weight average for the childs age),second degree (6175%), and third degree (60% and less).Their article linked this classification system to theprecise health outcomemortality and assignedto varying degrees of malnutrition not only a clinicalvalue but also a prognostic significance. The authorsdocumented that the type of prognosis dependedmainly on the severity of malnutrition, measured asweight deficit. Subsequently, reference to first, secondand third degrees of malnutrition became commonjargon not only among nutritionists, but also amongothers working in the field of child health. With time,the so-called Gomez classification (using theHarvard reference values (3) and different cut-offpoints, i.e., 80%, 70% and 60% of median) was usedwidely both to classify individual children for clinicalreferral and to assess malnutrition in communities.

The paper by Gomez et al. raised twointerrelated issues that are discussed below. The

first, which describes how to measure malnutrition,considers such methodological issues as selectinganthropometric indicators, choosing reference dataand establishing cut-off points. The second issueconcerns the relationship between malnutrition, asmeasured by child anthropometry, and mortality.

Measuring nutritional status

The classification developed by Gomez et al. wasbased on three prior selections: an anthropometricindicator, a reference population with which tocompare the index child or community, and cut-offpoints to classify children according to variabledegrees of malnutrition. Classifications developedafter Gomez have all relied on these same threeelements.

Anthropometric indicatorNutritional status can be assessed using clinical signsof malnutrition, biochemical indicators and anthro-pometry. Inadequacies in nutritional intake even-tually alter functional capacity and result in manyadverse health outcomes that are distinct expressionsof malnutritions different levels of severity. Initially,children adapt to inadequate diets through reducedphysical activity and slowed rates of growth. Atmoderate degrees ofmalnutrition activity and growthrates are affected to a greater degree and, in addition,signs of wasting and some biochemical abnormalities(e.g. reduction in serum albumin) begin to show. Atadvanced stages of severity, all linear growth ceases,physical activity is severely curtailed, body wasting ismarked, and clinical signs (e.g. oedema, hair and skinchanges) are noticeable. Anthropometry thus has animportant advantage over other nutritional indica-tors: whereas biochemical and clinical indicators areuseful only at the extremes of malnutrition, bodymeasurements are sensitive over the full spectrum. Inaddition, anthropometric measurements are non-

1 Medical Officer, Department of Nutrition for Health andDevelopment, World Health Organization, 1121 Geneva 27,Switzerland (email: [email protected]).

Ref. No. 00-0826

1271Bulletin of the World Health Organization, 2000, 78 (10) # World Health Organization 2000
Alejandro MartinezResaltadoAlejandro MartinezResaltadoAlejandro MartinezResaltadoAlejandro MartinezResaltado

invasive, inexpensive and relatively easy to obtain.Themain disadvantage of anthropometry is its lack ofspecificity, as changes in body measurements are alsosensitive to several other factors, including intake ofessential nutrients, infection, altitude, stress andgenetic background.

A childs body responds to malnutrition in twoways that can be measured by anthropometry: adeceleration or cessation of growth, which over thelong term results in low height-for-age or stunting;and body wasting, which is a short-term response toinadequate intakes, and commonly assessed byweight relative to height. Height-for-age andweight-for-height thus discriminate between dif-ferent biological processes, unlike weight-for-age,which could be low because of stunting (shortstature) and/or wasting (recent weight loss). TheGomez criteria relied exclusively on weight-for-ageand hence could not discriminate between short-term and long-term forms of malnutrition. Thus,patients classified on the basis of weight-for-agecriteria are a mixed group in terms of their clinicalnutritional status. In post-Gomez classifications,weight-for-height has emerged as a very importantindicator (4, 5) and, in fact, several authors haveidentified low weight-for-height as the indicator ofchoice for screening severely malnourished childrenwho are at increased risk of dying (69).

Reference populationAnthropometric values are compared across indivi-duals or populations in relation to a set of referencevalues. The choice of reference population to assessnutritional status has a significant impact on theproportion of children identified as being malnour-ished and, in turn, important programmatic implica-tions for what to do about it (10). Much has beenwritten about growth references, but there remainunanswered questions about the many factors thatdetermine human growth and indeed what constitu-tes normal growth. Gomez et al. used theoreticalweights among Mexican children (2), and laternutrition classification standards have followed thistradition by choosing reference values of their own. Adetailed account of the different growth referencesused prior to the current international reference isprovided elsewhere (11). The USNational Center forHealth Statistics (NCHS)/WHO international refer-ence, in use since the late 1970s, has been found tohave important technical and biological drawbacks.Consequently, WHO is conducting a multicountrystudy aimed at developing a new growth reference. Amajor innovation of this new effort is the use of aninternationally constituted reference population asopposed to the strictly national samples in existingreferences (12). The extent to which the new curvesdiffer from the current ones in shape and the spreadof values around the mean will affect the relationship established using the old reference values between child anthropometry and functional out-comes such as mortality.

Cut-off pointOnce an anthropometric indicator and a referencepopulation have been selected, it is necessary todetermine the limits of normality. There are threeclassification systems for comparing a child, or agroup of children, to the reference population: Z-scores (standard deviation scores), percentiles andpercent-of-median. The Gomez classification usesthe percent-of-median, which is a convenientmeasure if the reference population distribution hasnot been normalized. The percent-of-median issimpler to calculate than a Z-score or percentile. Inthe growth reference populations used prior to theNCHS/WHO reference, the curves were generallynot normalized. However, in order to formulate thesoftware version of the current reference, the originalheight and weight distributions were slightly mod-ified by a normalization procedure (13). Since thecalculation of the percent-of-median ignores thedistribution of the reference population around themedian, the interpretation of any given percent-of-median value varies across age and height groups. Forexample, depending on a childs age, 80% of themedian weight-for-age can be above or below -2 Z-scores, resulting in different classifications of healthrisk. In addition, common cut-offs for percent-of-median are different for the three distinct anthropo-metric indicators (5).

Since the late 1970s WHO has recommendedusing the Z-score system because of its severaladvantages (4). For population-based applications,the software version of the NCHS/WHO referencegreatly contributed to the wide acceptance of the Z-score concept because it simplified the handling ofanthropometric data obtained from surveys andnutritional surveillance. For individual applications,however, there has been reluctance to adopt itbecause the Z-score of an individual child is moredifficult to calculate than the percent-of-median.While field staff generally have no difficulty learninghow to perform the calculation, they frequentlyexperience difficulties with understanding the con-cept of the Z-score. It is nevertheless generallyrecognized that Z-score is the most appropriatedescriptor of nutritional status for both individualand population-based applications, and health andnutrition centres are gradually switching to its use.Teaching how to use Z-scores, however, remains achallenge, and imaginative and simple ways need tobe developed to convey this concept to healthprofessionals.

The use of a statistically defined cut-off point(e.g. -2 Z-score) is not unique to anthropometry;indeed, it is widely applied in many clinical andlaboratory tests. Nevertheless, it is important to bearin mind that using a cut-off-based criterion to definewhat is abnormal is somewhat arbitrary. In reality,there are not two distinct populations one well-nourished and the othermalnourishedbut rather acontinuous gradation of nutritional status. That is,the risk of undesirable health outcomes such asmortality does not change dramatically by simply

1272 Bulletin of the World Health Organization, 2000, 78 (10)

Alejandro MartinezSubrayadoAlejandro MartinezSubrayadoAlejandro MartinezResaltadoAlejandro MartinezResaltadoAlejandro MartinezResaltado

crossing the cut-off line: significant deteriorationwithin the normal range may in fact carry greaterrisk. For many purposes, the best descriptor of apopulations nutritional status is the mean, which inless developed environments is usually shifted to theleft. This population approach resolves theproblem of focusing solely on the severely mal-nourished subpopulation falling below a certain cut-off. In most instances, the mild and moderatelymalnourished subpopulations will be of greaterimportance from a public health perspective becausethere are many more children here than in theseverely malnourished category.

Predicted risk, which drives most interventions,focuses on individuals, where the farther away from thecentre of the distribution an individual is, the greaterthe risk of outcomes such as mortality. However, it isinadequate for nutritional interventions to be drivensolely by an individual approach, limiting nutritionalsupport to children who fall below the accepted cut-offlevel. This approach tackles only the tip of themalnutrition iceberg. Ideally, both the population andindividual approaches should be combined so thatchildren who remain severely malnourished despitepopulation-based interventions are identified andgiven special therapeutic attention (9).

Present practice often recommends the use ofa universal cut-off point, e.g. -2Z-score, which is veryuseful for population-based monitoring. However,for individual applications in screening high-riskchildren, cut-off points should be locally identified bytaking into account: the population-specific preva-lence and nature of malnutrition; the cut-off pointbelow which children are shown to respond tospecific interventions; and the availability of re-sources, which will ultimately determine the propor-tion of children that the intervention can reach.

Child anthropometry and mortality

The pointmade by FedericoGomez and colleagues (1) that severe malnutrition has a significant effect onmortality is biologically plausible and hardly everdisputed. Several other studies have documented thatseverely malnourished children are at a much greaterrisk of dying than are healthy children (14). An equallyimportant question is, how strong is the associationbetween mild or moderate malnutrition and the risk ofchild mortality? An accurate answer is important forthe success of child survival programmes as thenumber of children with mild and moderate malnutri-tion is several times greater than the number who areseverely malnourished (15). If mild and moderatemalnutrition are strongly associated with increasedmortality, efforts to reduce child mortality should bedirected to improving the nutritional status of allchildren, instead of focusing primarily, or exclusively,on severely malnourished patients.

Few large prospective studies of mortalityduring childhood have examined this issue. The oneby Chen et al. (16), who studied a cohort ofBangladeshi children (1526 months at enrolment)for two years, has been highly influential. Theirobservations, which had important programmaticimplications, showed a pronounced threshold effect:mortality increased with worsening nutritional statuswhen malnutrition was severe, but mild or moderatedegrees of malnutrition had little predictive power.More recently, Pelletier et al. (17, 18) reviewed28 community-based prospective studies on therelationship between anthropometric indicators ofmalnutrition and childmortality. The authors reachedtwo important conclusions. First, the accumulatedresults were consistent in showing that the risk ofmortality was inversely related to anthropometricindicators of nutritional status and that there was anelevated risk even at mild-to-moderate levels ofmalnutrition. Moreover, when considering the re-lative proportions of severe versus mild-to-moderatemalnutrition in populations, the authors showed thatthe majority of nutrition-related deaths were asso-ciated with mild-to-moderate, rather than severe,malnutrition. In programmatic terms, this impliesthat strategies focusing primarily or exclusively onseverely malnourished children will be inadequate toimprove child survival in any significant way. Tomake a substantial impact onmortality, the burden ofmild and moderate malnutrition in a population mustalso be reduced. The second important result fromthe review by Pelletier et al. is the confirmation thatmalnutrition has a potentiating (multiplicative) effecton mortality. Malnutrition, rather than acting in asimple additive fashion, was in fact observed tomultiply the number of deaths caused by infectiousdisease.

The substantial contribution to child mortalityof all degrees of malnutrition is now widelyrecognized. As a consequence, current internationalefforts such as the Integrated Management ofChildhood Illness strategy, which focuses on themost important causes of child death, include anumber of key nutritional interventions (19). It waspioneers like FedericoGomez and his colleagueswholaid the groundwork for todays approach bydeveloping the concepts that the internationalnutrition community now takes for granted andcontinues to refine in an effort to understand betterthe magnitude of malnutrition and its impact onhealth. Those who believe that assessing nutritionalstatus is a fundamental tool for protecting childhealth are indebted to this pioneering work. n

AcknowledgementI would like to thank Dr Adelheid Onyango forher useful comments on the first draft of thiscommentary.

1273Bulletin of the World Health Organization, 2000, 78 (10)

Measuring nutritional status

References

1. Gomez F et al. Mortality in second and third degree malnutrition.Journal of Tropical Pediatrics, 1956, 2: 7783.

2. Gomez F et al. Malnutrition in infancy and childhood with specialreference to kwashiorkor. In: Levine SZ, ed. Advances in pediatrics.New York, Year Book Publishers, 1955, VII: 131169.

3. Stuart HC, Stevenson SS. Physical growth and development.In: Nelson WE, ed. Textbook of Pediatrics. 5th ed. Philadelphia,WB Saunders, 1950: 1473.

4. Waterlow JC et al. The presentation and use of heightand weight data for comparing nutritional status of groupsof children under the age of 10 years. Bulletin of the WorldHealth Organization, 1977, 55: 489498.

5. Physical status: the use and interpretation of anthropometry.Report of a WHO Expert Committee. Geneva, World HealthOrganization, 1995 (WHO Technical Report Series, No. 854).

6. Bern C et al. Assessment of potential indicators forprotein-energy malnutrition in the algorithm for integratedmanagement of childhood illness. Bulletin of the WorldHealth Organization, 1997, 75: 8796.

7. van den Broeck J, Meulemans W, Eeckels R. Nutritionalassessment: the problem of clinical-anthropometrical mismatch.European Journal of Clinical Nutrition, 1994, 48: 6065.

8. Trowbridge FL. Clinical and biochemical characteristicsassociated with anthropometric categories. American Journalof Clinical Nutrition, 1979, 32: 758766.

9. Management of severe malnutrition: a manual for physicians andother senior health workers. Geneva, World Health Organization,1999.

10. WHO Working Group on Infant Growth. An evaluationof infant growth: the use and interpretation of anthropometryin infants. Bulletin of the World Health Organization, 1995,73: 165174.

11. de Onis M, Yip R. The WHO growth chart: historicalconsiderations and current scientific issues. In: Porrini M, Walter P,eds. Nutrition in pregnancy and growth. Basel, Karger, 1996(Bibliotheca Nutritio et Dieta, 53: 7489).

12. WHO Working Group on the Growth Reference Protocoland WHO Task Force on Methods for the NaturalRegulation of Fertility. Growth patterns of breastfed infantsin seven countries. Acta Paediatrica, 2000, 89: 215222.

13. Dibley MJ et al. Development of normalized curves forthe international growth reference: historical and technicalconsiderations. American Journal of Clinical Nutrition, 1987,46: 736748.

14. Schofield C, Ashworth A. Why have mortality rates for severemalnutrition remained so high? Bulletin of the World HealthOrganization, 1996, 74: 223229.

15. de Onis M, Blossner M. WHO Global Database on Child Growthand Malnutrition. Geneva, World Health Organization, 1997.

16. Chen LC, Chowdhury AKMA, Huffman SL. Anthropometricassessment of energy-protein malnutrition and subsequent riskof mortality among preschool aged children. American Journalof Clinical Nutrition, 1980, 33: 18361845.

17. Pelletier DL, Frongillo EA, Habicht JP. Epidemiologic evidencefor a potentiating effect of malnutrition on child mortality.American Journal of Public Health, 1993, 83: 11301133.

18. Pelletier DL. The relationship between child anthropometryand mortality in developing countries: implications for policy,programs and future research. Journal of Nutrition, 1994,124: 2047S2081S.

19. Tulloch J. Integrated approach to child health in developingcountries. Lancet, 1999, 354 (suppl. II): 1620.




MORTALITY IN SECOND AND THIRD DEGREE MALNUTRITION*

by

FEDERICO GOMEZ, M.D., RAFAEL RAMOS GALVAN, M.D., SILVESTRE FRENK, M.D., JOAQIUN CRAVIOTO Mu:Noz, M.D., RAQUEL Cl!AvEz, M.D. AND JumTH VAzQUEZ, M.D.

(Nutrition Department, Hospital Infantil de Mexico, Mexico City.)

Malnutrition has been defined as a pathological condition of varying degrees of severity, and diverse clinical manifestations, resulting from the deficient assimilation of the components of the nutrient complex (G6MEz, 1955). This disease affects the physico-chemical pattern of the tissues, reduces the defensive capacity to environmental aggressions, lowers both the efficiency and the ability for work, and shortens life (EscUDERO, 1935 ; TROWELL, 1948 ; DAVIES, 1952 ; ZUBIRAN, 1953).

The disease attacks with greater intensity certain social groups, and has a considerable clinical importance.during critical stages of development of the child, such as infancy and the pre-school age, which are characterized by rapid growth and high nutritional needs.

The causes responsible for malnutrition may be classified as primary- insufficient food supply, or under-nutrition ; and secondary or conditioned- poor absorption, increased excretion, increased requirements (JoLLIFFE, 1950). Most of the patients that come to our Department are children suffering from chronic underfeeding resulting from an insufficient diet, both in quantity and quality, to which they have been submitted for at least three-quarters of their lives.

When underfeeding is moderate, or has acted for only a short time, the " nutritional reserves " of the organism are only partially depleted, and malnutrition exhibits a mild clinical picture, where the body weight ranges between 76-90 per cent. of the theoretical average for the child's age. This, we call first degree malnutrition. As the effect of under-feeding becomes more serious, the picture becomes more marked, resulting in second degree malnutrition. At this stage, the weight is between 61-75 per cent. of the theoretical average for the age. The clinical picture, the prognosis, and the treatment become much more complicated, and frequently the patient requires hospital care.

In third degree malnutrition, when the nutritional reserves are practically exhausted, the maximum weight is never more than 60 per cent. of the average for the age ; while, in addition, there are serious somatic and functional, including psychological, changes. Treatment is very complicated and expensive, and the patient must be hospitalized. At this stage, the disease has a high mortality rate - figures given by several authors ranging from 30 to 60 per cent. (LEVINDER, 1912 ; GILLMAN, 1951 ; DEAN, 1954).

This classification of the disease - according to its varying degrees of severity and its various clinical manifestations- which has been adopted by us since 1946, has often-times been criticised. Recently the Guatemalan research workers have introduced the so-called "concept of Incipient Infantile Pluricarential Syndrome (pre-Kwashiorkor)" (SCRIMSHAw, 1955). In other words, they also accept degrees of severity, but use a different terminology.

We have conducted a careful study of both the clinical picture preceding death and of the apparent cause of death in a large group of children suffering from chronic malnutrition due to underfeeding seen between 1949 and 1952, and useful information has thereby been obtained which will assist in assessing the prognosis and the correct form of treatment in this type of case in the future. In due course, we shall also report on a second group studied from 1953 through 1955.

MATERIAL

Our study consisted of 733 children admitted to the Nutrition Department of the Hospital Infantil de Meldco, from 1949 through 1952. These cases had the following characteristics : (a) Previous diet. All the children. had been submitted to a severe and prolonged restriction of food (Table 1). The sources of protein were corn, beans, and wheat in small amounts. This diet was deficient in lysine, tryptophane, isoleucine, valine, threonine, methionine and cystine. The biological value of the protein mixture, calculated according to MITCHELL and BLOcK's formula (1946), was only 69 per cent. Besides, it was consumed only in small amounts, consisting of 50 per cent. of the essential caloric, and 20-60 per cent. of the normal protein, requirements. (b) Sex incidence. 48 per cent. were males and 52 per cent. females. (c) Average age. 31 17 months. (d) Weight. 52 lO per cent. of the normal weight for the age (Table II). (e) Height. This was less affected than weight, with maximum variations of 15 per cent. (f) Clinical oedema. Present in 71 per cent and absent in 29 per cent. (g) Skin lesions The incidence of these is shown in Table III. (h) Stigmata of malnutrition. Other signs of malnutrition are shown in Table IV.

* Studies on the Undernourished Child No. XIV.

The Journal of Tropical Pediatrics, September, 1956


METHODS

Mortality was studied taking into consideration the time that elapsed between ad-mission and death of the patient. Through the information obtained from the clinical history and the physical examination, attempts were made to determine the cause of death

TABLE I. Composition of the diet previous to hospitalization (GoMEZ, 1952).

Daily intake Percentage of normal

Calories 700- 800 40- 50 Protein 10- 30 gm. 20-60 Fat 5- 15 gm. 10- 33 Carbohydrate 80- 140 gm. 60- 85 Calcium 100 - 400 mgm. 10-40 Phosphorus 200 - 500 mgm. 20-50 Iron 4- 8mg. 50-80 Vitamin A* 2000 I.U. -Thiamine* 400 - 1200 gamma Sufficient Riboflavin* 500 - 1500 mgm. Low Nicotine acid* 1- 2.5 mgm. Low Ascorbic acid* Less than 50 mgm. Low

* Vitamins calculated in raw foods.

TABLE II. Distribution of body weights in terms of percentage of the theoretical weight for the age.

Weight No. of cases Percentage

31 to 40% of the normal 75 10.50 41 to 50% ,

" " 241 33.75

51 to 60% , " "

268 37.53 61 to 70% ,

" " 105 14.70

71 to 80% , " "

20 2.80 81 to 90% ,

" " 3 0.42

91 to 100 , " "

2 0.28

in each case. The significance of some positive or negative signs due to malnutrition or to other causes was also determined.

The data obtained were submitted to routine statistical analysis (MAINLAND, 1952) ; a variant of the usual formula to calculate " X2 " was used to estimate the possible influence of certain clinical situations on mortality.

TABLE III. Incidence of skin lesions.

Without With clinical clinical

Type of skin lesion oedema oedema "t "* (Percentage) (Percentage)

Dry hyperchromic skin 90 94 Very dry hyperchromic with mosaic

appearance 38 46 Follicular hyperkeratosis 20 26 Hyperkeratosis palmaris et plantaris 48 37 Fissures 3 8 Seborrhoea 42 49 Pellagrous erythema 13 47 4.94 Acute pellagrous dermatitis 18 40 2.94 Dyskeratotic hyperchromic lesions 50 83 3.00 Desquamating lesions in large flaps 8 21 2.48 Desquamating lesions in small flaps 13 37 3.81 Postdesquamation hypochromia 15 26 Hyperchromia along capillary circulation 1.4 8 3.26 Crusty lesions suggesting post-purpuric

lesions 6 11 Purpuric lesions 24 28 Perifolliculosis 1.4 11 3.07 Coldness and cyanosis of hands and feet 62 82 Marblization 11 18 Telangiectasis 1.4 6 Gangrenous lesions and eschars 11 18 Hypertrichosis 24 20 "Wet cloth" sign 27 11 Abdominal superficial circulation 6 7

* " t " shows the significant degree of difference between children with and without clinical oedema. " t " values below 2 were not reported, as they are not significant.



TABLE IV. Certain clinical features and laboratory investigations.

Diarrhoea Dehydration Parental infection Yomiting Fever Hypothermia

Intestinal parasites Stool cultures Shigellae Salmonellae Other agents Mantoux test Serological test for syphilis

REsULTS

(1) Mortality.

There were 234 deaths in the 733 cases studied, which gives a mortality of 31 per cent. Fig. 1 shows the number of deaths that occurred during the first 14 days.

Ill w

70

eo

so

~40 ... 0 Ill:

During the first 48 hours, there ~ 30 were 105 deaths (44 per cent.). There ~ was a definite difference between the z 20 mortality rate on the first day (29 per cent.), and on the second day (15 per cent.). Eliminating the deaths which occurred during the first 48 hours, the mortality is reduced to 129 deaths out of 629 cases (20 per cent.).

(2) Probable causes of death.

10

0 2

Percentage positive

70 56 51 39 32 31

28 28 14.3 9.4 4.5 4 3

FIG. 1. Number of deaths observed during the first two weeks after admission in 733 cases of malnutrition

4

(Gomez, 1955).

e a 10 DAVS ELAPSED

12 14

Water and electroyte imbalance was evident in 48 children out of 69 who died within the first 24 hours. Acute broncho-pneumonia, either alone or combined with water and electrolyte imbalance, was responsible for death in another group of 35. On the second day, the mortality rate dropped to 36- out of these, there were 21 with disturbance of water and electrolyte metabolism and 18 with acute bronchopheumonia, or a combination of both.

Between the third and the seventh days, there were 76 deaths, and again electrolyte and fluid imbalance (50 cases) alone or combined with acute bronchopneumonia (35 cases), was considered to be responsible (Table V).

TABLE V. Distribution of mortality by weeks.

Weeks 1st. 2nd 3rd. 4th. Late cases

Number of cases 181 29 10 7 7

With fluid and electrolyte imbalance 119 20 5 3 0

With bronchopneumonia 88 14 5 6 3

Fluid and electrolyte im-balance alone 47 4 2 0 0

Combined with broncho-pneumonia 52 8 3 2 0

Fluid and electrolyte im-balance with other causes 20 8 0 1 0

Bronchopneumonia with other causes 9 1 2 2 3

Bronchopneumonia alone 27 5 0 2 3 Other causes 26 3 3 0 4

September, 1956, The Journal of Tropical Pediatrics


(3) Effect on mortality of certain clinical features found on admission.

(a) Weight. The death rate in 120 cases of second degree malnutrition was 22.6 per cent.; while in 544 cases of third degree malnitrition, it reached 33.53 per cent. The significant statistical difference is : t = 2.3 (Table VI).

(b) Clinical Oedema. The gross mortality in patients with clinical oedema was 29.67 per cent., and in those without this sign it was 35.98 per cent. -the difference was not statistically significant. On further analysis of these cases by the elimination of deaths that occurred within the first 48 hours, the mortality was found to be 20 per cent. and 22.6 per cent. respectively, with "t" = 0.74 (Table VI).

(c) Skin lesions of "pellagra." The statistical significance on mortality of cases with lesions of " pellagra," may be seen also in Table VI.

TABLE VI. Effect on mortality of certain signs of malnutrition found on admisiosn.

Clinical groups No. of Total Early mortality Late mortality cases mortality (1st--48 hrs.)

Percent- Percent- Percent-age "t ''* age " t "* age "t "*

deaths deaths deaths

Second degree malnutrition 128 22.65 5.47 18.18 2.30 2.80 0.80

Third degree malnutrition 584 33.53 14.70 21.50

With clinical oedema 519 29.67 12.13

I

20.00 1.67 1.85 0.74

Without clinical oedema 214 35.98 17.29 22.60

With " Pellagra " 465 31.70 12.25 18.60 0.09 1.62 0.93

Without " Pellagra " 268 31.40 16.05 21.80

* A value of " t " above 2 shows significant differences.

According to these data, in our clinical material the groups with and without pellagra, and with and without oedema, showed no difference in mortality ..

(4) Effect on mortality of certain clinical features not due to malnutrition found on admission.

(a) Fluid and electrolyte imbalance. Table VII demonstrates the differences in mortality between children suffering from an obviously upset water and electrolyte balance on admission, and those who did not. The presence of this type of disturbance has an obvious effect on the mortality rate.

(b) Diarrhoea. The possible influence of this symptom was also considered and it was studied in children without fluid and electrolyte imbalance, but with diarrhoea on admission. Table VII shows that the difference in mortality rate between children with and without this symptom is significant, though less so than when comparing the presence and absence of dehydration.

(c) Bronchopneumonia. The differences in mortality were analysed only in those cases where the cause of death was an acute pulmonary process, present since the time of admission, but without a clinically evident fluid and electrolyte imbalance (Table VII). The difference between both groups is significant.



TABLE VII. Effect on mortality of some clinical signs found on admission, not directly due to malnutrition.

Early mortality Total mortality (1st. --48 hrs.) Late mortality

Clinical groups No. of cases Death Death Death

percent- " t "* percent- " t ,,.. percent- " t "* age age age

With fluid and electrolyte imbalance 411 44.00 19.10 31.53

7.98 12.00 12.00 Without fluid and electric im-

balance 322 15.00 7.10 9.36

With diarrhoea (but without fluid imbalance) 152 21.40

Without diarrhoea or fluid 2.60 imbalance 168 10.70

With bronchopneumonia, but without fluid imbalance 39 53.84

4.86 Without bronchopneumonia

or fluid imbalance 288 11.08

* A value of " t " above 2 shows significant differences.

DISCUSSION

As pointed out by DEAN (1954) and emphasized by BROCK(l9~5), the accurate compar-ison of mortality rates in cases of infantile malnutrition is difficult. Nevertheless, the high mortality found in our department agrees with that reported by GILLMAN and GILLMAN (1951), TRowELL (1954) and many others. The first mentioned authors point out that gross mortality varies from one year to another, fluctuating from 30 to 50 per cent., " without any objective index of the gravity of the acute episode." It is obvious that in our series water and mineral imbalances, and acute pulmonary processes, have a definite influence on gross mortality, and this influence extends throughout the first two weeks of hospital stay.

Independently of the cause of death, differences in weight have a significant effect on mortality. There is a marked difference in mortality during the first 48 hours between children with second degree malnutrition, and those with third degree. With these findings, the classification of malnutrition by degrees acquires, not only a clinical value, but a definite prognostic significance as well.

It has also been ascertained that other pathological disturbances, often found in severe malnutrition, such as oedema and "acute" skin lesions, have no influence on mortality. In view of the tendency other authors still have to consider as different entities cases with or without clinical oedema and " acute " skin lesions, the above findings show that there is probably no sound clinical basis for such classification.

In contra-distinction to the above, it has been shown that the difference in mortality in children with clinically evident water and mineral imbalance, and in those without it, gives a very high" t" (7.98), "p" being> 0.001. Similarly-, when diarrhoea was present on admission, it also influenced mortality, possibly through the production of water and mineral imbalance, which becomes established much more easily in children with serious malnutrition than in healthy ones. There is evidence that malnutrition per se may be

September, 1956, The Journal of Tropical Pediatrics


responsible for a negative potassium balance (HANSEN, 1954), and this could partly explain why diarrhoeal processes, however moderate, can bring about a marked derangement in water and mineral metabolism in these children, and why successful treatment of this situation is so difficult. The possible influence of the severity of undernutrition upon this situation is illustrated by the higher early mortality in third grade malnutrition.

The statistical analysis of the cases of acute bronchopneumonia without clinically evident fluid and water imbalance is especially difficult. The diagnostic signs, as pointed out by TROWELL et al (1954) may be misleading. Moreover, bronchopneumonia by itself may be responsible. for the production of acid-base disturbances, which may not be clinically apparent. On the other hand, a process of this type may be the sole cause of death by interfering with respiratory functions.

Malnutrition in itself may be regarded as a predisposing factor, but not as the direct cause of such a high mortality. An answer to this acute problem may perhaps be sought in a better understanding of the patterns of water and electrolyte metabolism in malnutri-tion, which obviously are different from those described in well-nourished children.

SUMMARY

The mortality rate in 733 malnourished children hospitalized from 1949 to 1952 has been subjected to analysis. The significant influence on mortality of the degree of mal-nutrition and the presence of water and mineral imbalance, diarrhoea and acute broncho-pneumopathy has been demonstrated. The existence of evident oedema or of skin lesions has no influence on mortality. The significance of these findings has been discussed.

REFERENCES

Brock, J. F., Hansen, J. D. L., Howe, E. E., Pretorius, P. J., Dave!, J. G. A., Hendrickse, R. G. (1955). Lancet, 2, 355, Kwashiorkor and Protein Malnutrition.

Dean, R. F. A., Schwartz, R. (1954). Courier, 4, 292. Protein Malnutrition in Infants. Davies, J. N. P. (1952). Ann. Rev. Med. 3, 99. Nutrition and Nutritional diseases. Stanford, Annual

Reviews, Inc. Escudero, P. (1935). Trabajos y Publicaciones del I. Municipal de Ia Nutricion. 1, I. Las !eyes de Ia

Nutricion. Gillman, T. and Gillman, J. (1951). Perspectives in Human Malnutrition. New York : Grune and

Stratton, Inc. Gomez, F., Ramos, G. R. and Cravioto, J. (1952). Pediatrics. 10, 513. Nutritional Recovery Syndrome.

- Preliminary Report. --,---,---and Frenk, S. (1955). Advances in Pediatrics. Vol. VII, p. 131. "Malnutrition

in Infancy and Childhood with special reference to Kwashiorkor." Chicago, Ill.:-Year Book Publishers Inc.

Hansen, J. D. L., Brock, J. F. (1954). Lancet, 2, 477. Potassium deficiency in the pathogenesis of nutritional oedema in infants.

Jolliffe, N., Tisdall, F. F., Cannon, P. R. (1950). Clinical Nutrition. New York, Paul B. Hoeber, Inc. Levinder, C. H. (1912). Pup. Health Rep. 27, 2076. The Prevalence and Geographic Distribution of

Pellagra in the United States. Mainland, D. (1952). Elementary Medical Statistics p. 1952. Philadelphia : W. B. Saunders & Co. Mitchell, H. H., Block, R. J. (1946). J. bioi. Chem., 163, 599. Some Relationships between the Aminoacid

Contents of Proteins and their Nutritive Value. Scrimshaw, N. S., Behar, M., Perez, C. Viteri, F. (1955). Pediatrics, 16, 378. Nutritional Problems of

Children in Central America and Pamana. Trowell, H. C. (1948). Trans. R. Soc. trap. Med. Hyg., 42, 417. Malignant malnutrition (Kwashiorkor). --,Davies, J. N. P. and J;>ean, R. F. A. (1954). "Kwashiorkor" London : E. Arnold Ltd. Zubiran, S. and Gomez-Mont, F. (1953). Vitamins and Hormones. -Vol. XI, p. 97. Endocrine

Disturbances in Chronic Human Malnutrition. New York : Academic Press Inc.


Documents

Measuring nutritional status in relation to mortality