Kwashiorkor and Marasmus: Distinguishing Severe Acute Malnutrition in Clinical Practice

In a nutrition stabilisation centre in Juba, South Sudan, during the food crisis of 2017, a paediatric nurse records two admissions within the same hour. The first child, aged twenty-two months, has skin that peels in sheets across his trunk, hair that has turned reddish and pulls out in tufts, and bilateral pitting oedema that extends to his knees. He is irritable, withdrawn, and refuses to eat. The second child, eleven months old, is skeletal - her ribs are countable from across the ward, her thighs barely larger than a clenched fist, her face the hollow, aged appearance that clinicians describe as “old man facies.” Her skin clings to her bones without oedema, and unlike the first child she is alert, though visibly distressed. Both children have severe acute malnutrition. Both will require weeks of structured nutritional rehabilitation. But they do not have the same condition, and conflating them - diagnostically or therapeutically - carries measurable risk of preventable death.

This distinction is not merely academic. In the countries of Sub-Saharan Africa where severe acute malnutrition (SAM) reaches its highest global burden, clinicians operating in under-resourced facilities must routinely differentiate kwashiorkor from marasmus on the basis of physical examination alone, often without laboratory support. Understanding the clinical signatures, pathophysiological mechanisms, and syndrome-specific implications for treatment is among the most consequential skills in frontline paediatric nutrition.

Epidemiological Context: The Burden of SAM in Sub-Saharan Africa

Globally, approximately 19 million children under five suffer from SAM at any given time, the overwhelming majority in Sub-Saharan Africa and South Asia.¹ Black et al. (2013), writing in The Lancet, estimated that undernutrition in all its forms - including SAM - contributes to 45% of all deaths in children under five, or approximately 3.1 million deaths annually.² The burden is not evenly distributed: case rates are highest in the Sahel belt, the Horn of Africa, and the Great Lakes region, where structural food insecurity, high infection burdens, and weak health systems converge with lethal regularity.

Country-specific data confirm the scale of the problem. In Niger, acute malnutrition prevalence in children under five reached 14.8% during the 2012 Sahel crisis, with SAM affecting an estimated 322,000 children - approximately one in seven in the affected zones. In the Democratic Republic of Congo, the SMART survey of 2018 recorded a global acute malnutrition prevalence of 12.1% nationally, rising to 21.9% in the Kasaï provinces. In South Sudan, the 2017 Integrated Food Security Phase Classification placed over one million children under five in acute nutritional emergency, with SAM case-fatality rates in under-resourced facilities exceeding 20% without structured therapeutic intervention. Nigeria, with the largest absolute child population in Sub-Saharan Africa, contributes disproportionately to global SAM burden: a pooled analysis of National Nutrition and Health Surveys between 2010 and 2018 recorded wasting prevalence consistently above 7% nationally, reaching 18% in the north-western zones of Sokoto and Zamfara.

Case fatality rates within this population are not fixed. Pelletier et al. (1995) demonstrated that even mild-to-moderate undernutrition substantially potentiates child mortality from infectious disease, with the relationship following a synergistic rather than additive model - a finding that challenges the clinical tendency to reserve intensive intervention for the most extreme presentations.³ Guerrant et al. (2008) elaborated this nexus further, documenting how enteric infections accelerate intestinal mucosal injury and nutrient malabsorption in already-depleted children, creating a self-reinforcing cycle that is clinically recognisable as the malnutrition-infection spiral.⁴

For those interested in the surveillance infrastructure that enables such national estimates, our overview of implementing Health and Demographic Surveillance Systems addresses the methodological underpinnings in detail.

Defining the Two Syndromes

Kwashiorkor: Definition, Origins, and Clinical Presentation

The term kwashiorkor derives from the Ga language of coastal Ghana, where it was used to describe the condition observed in a child displaced from the breast by a new sibling - “the disease the deposed child gets.” Cicely Williams, working in the Gold Coast in the 1930s, was the first to document it systematically as a distinct clinical entity characterised by oedema, dermatosis, and hair changes in children whose diets were predominantly starchy.

The kwashiorkor definition most widely accepted in current clinical practice refers to severe acute malnutrition characterised by bilateral pitting oedema extending at least to both ankles, in the absence of an obvious non-nutritional cause. Oedema is the defining diagnostic criterion: its presence, regardless of weight-for-height z-score (WHZ), classifies a case as kwashiorkor or marasmic-kwashiorkor (when wasting and oedema co-exist). The oedema of kwashiorkor is typically dependent, beginning in the feet and ankles and progressing proximally to involve the legs, abdomen, and - in severe cases - the face, producing the characteristic moon face appearance that can mislead a cursory examiner into overestimating the child’s weight and nutritional status.

Beyond oedema, the clinical features of kwashiorkor include:

• Dermatosis: A pathognomonic skin condition known as “flaky paint” or “crazy paving” dermatosis, characterised by hyperpigmented patches that desquamate, leaving hypopigmented, friable skin beneath. It is most common over pressure areas and the lower limbs, and constitutes an independent risk factor for sepsis due to loss of the skin’s barrier function.
• Hair changes: Depigmentation, straightening, and easy pluckability of scalp hair are characteristic, and the “flag sign” - alternating bands of normal and depigmented hair - may be visible on close examination when the child has experienced repeated nutritional insults separated by partial recovery.
• Hepatomegaly: Fatty infiltration of the liver, resulting from impaired synthesis and export of very-low-density lipoproteins (VLDL), produces a palpably enlarged, non-tender liver in the majority of kwashiorkor cases.
• Apathy and anorexia: Children with kwashiorkor are frequently profoundly disinterested in food - a clinically treacherous feature, as therapeutic feeding depends on voluntary intake in the absence of nasogastric tube placement.
• Hypoalbuminaemia: Where laboratory capacity exists, serum albumin is typically below 30 g/L, reflecting the collapse of hepatic protein synthetic function.

Marasmus: Definition, Features, and Clinical Presentation

The marasmus definition refers to severe acute malnutrition characterised by severe wasting - a weight-for-height z-score below −3 of the WHO median (WHZ < −3) or a mid-upper arm circumference (MUAC) below 115 mm in children aged six to fifty-nine months - without bilateral pitting oedema. Marasmus represents the body’s adaptive response to prolonged caloric insufficiency: muscle mass and subcutaneous fat are catabolised systematically to maintain glucose supply to the brain and vital organs.

The clinical presentation is characterised by:

• Severe wasting: The child appears skeletal, with visible ribs, prominent clavicles, and wasted buttocks giving the appearance of “baggy pants” as skin folds hang loosely over diminished gluteal and thigh musculature.
• Old man facies: Loss of the buccal fat pads - among the last fat stores to be mobilised in healthy children - produces a hollow, prematurely aged facial appearance that is diagnostically distinctive.
• Preserved alertness: In contrast to kwashiorkor, marasmic children are typically alert and, notably, hungry - a prognostically relevant feature and a therapeutic advantage once feeding commences.
• Absent or minimal oedema: Albumin levels may be closer to the normal range than in kwashiorkor, because the body preserves circulating protein at the expense of somatic mass for longer in pure caloric restriction.
• Hypothermia and hypoglycaemia: Both are immediate threats requiring urgent management, particularly in settings with overnight temperature drops or prolonged fasting intervals.

The Intermediate Category: Marasmic-Kwashiorkor

A third clinical presentation - marasmic-kwashiorkor - combines features of both syndromes: wasting plus bilateral pitting oedema. This presentation tends to carry the highest case-fatality risk among the three SAM categories, reflecting both the depth of metabolic derangement and the clinical difficulty of managing concurrent fluid dysregulation and tissue wasting simultaneously. Manary and Sandige (2008) noted that approximately 14–20% of SAM admissions in Malawian facilities presented with marasmic-kwashiorkor, with mortality rates approximately double those of uncomplicated marasmus in the same cohort.⁵

Pathophysiology: Why the Syndromes Differ

The mechanistic divergence between kwashiorkor and marasmus has been debated for decades, and a fully unified explanation remains elusive. The classical teaching - that kwashiorkor results from protein deficiency in the presence of adequate caloric intake, whilst marasmus results from combined protein-energy deficiency - has been substantially revised by subsequent research. In practice, dietary protein intakes are often similarly low in populations where both syndromes co-exist, and the two conditions can present in children from the same household.

Current thinking emphasises the role of oxidative stress in the aetiology of kwashiorkor. Prentice et al. (2013), in a seminal paper on critical windows of stunting and wasting, argued that the oedema and dermatosis of kwashiorkor reflect overwhelmed antioxidant defences in the context of infection, aflatoxin exposure, and micronutrient depletion - particularly of zinc, selenium, and glutathione precursors.⁶ Children with marasmus, on this model, mobilise adaptive responses that attenuate oxidative damage despite severe caloric restriction; those who develop kwashiorkor fail to mount these defences, perhaps due to genetic variation in antioxidant enzyme activity or greater cumulative toxin burden.

The oedema itself reflects hypoalbuminaemia driving oncotic pressure reduction, sodium and water retention, and - in some cases - concurrent hormonal dysregulation including elevated cortisol and disrupted aldosterone activity. Hepatic dysfunction impairs the synthesis of clotting factors, carrier proteins, and enzymes critical to immune function, explaining the heightened susceptibility to septicaemia, hypoglycaemia, and electrolyte dysregulation that characterises kwashiorkor’s clinical course.

In marasmus, the pathophysiology is more legible: sustained negative energy balance triggers progressive catabolism of adipose and skeletal muscle tissue, suppression of growth hormone signalling, and reduction of basal metabolic rate. The gut undergoes atrophy of the intestinal villous architecture - a process that impairs absorptive capacity and contributes to the malabsorption of any nutritional rehabilitation introduced too rapidly, a phenomenon clinicians encounter as “refeeding diarrhoea.”

Diagnosis in Clinical Practice

Anthropometric Criteria

The diagnostic framework for SAM in children aged six to fifty-nine months rests on two primary anthropometric measures, either of which is sufficient for classification:

• Weight-for-height z-score (WHZ) < −3: Derived from the WHO 2006 Child Growth Standards, a WHZ below −3 SD indicates severe wasting. Measurement requires an accurate weight (using a tared scale, ideally digital) and height or length measurement with a stadiometer or length board. The logistical and training requirements of WHZ measurement, particularly in community and outreach settings, have driven increasing emphasis on MUAC as a primary screening tool.
• Mid-upper arm circumference (MUAC) < 115 mm: MUAC, measured with a graduated insertion tape at the midpoint of the left upper arm, has demonstrated both high sensitivity and high specificity for identifying children at imminent risk of SAM-associated death. Its operational advantages - requiring minimal training, no weighing equipment, and no age verification - make it particularly valuable for community health workers in under-resourced settings. Bhutta et al. (2013) affirmed MUAC as a scalable, cost-effective screening instrument within community-based management frameworks.⁷
• Bilateral pitting oedema: As noted above, oedema alone classifies a child as having SAM in the kwashiorkor category, irrespective of WHZ or MUAC. The clinical test requires firm pressure applied with both thumbs to the dorsum of both feet for three seconds; a visible pit remaining on release constitutes a positive finding.

Where laboratory facilities are available, investigations including blood glucose, serum electrolytes (particularly potassium and sodium), full blood count, and malaria rapid diagnostic testing (in endemic regions) should be obtained on admission, as the results directly inform the management of life-threatening complications.

Differential Diagnosis

Bilateral pitting oedema in a child can occasionally arise from non-nutritional causes - nephrotic syndrome, cardiac failure, or severe anaemia - and clinicians should be alert to features that suggest these diagnoses, including proteinuria, cardiac murmur, or haemoglobin below 4 g/dL. In practice, however, the co-presence of wasting, dermatosis, or hair changes in a food-insecure child from a SAM-prevalent setting makes kwashiorkor the working diagnosis until evidence demands otherwise.

Treatment Frameworks

The Phase Approach to Inpatient Management

The WHO-endorsed treatment protocol for complicated SAM, updated in 2013, divides inpatient management into three sequential phases designed to match nutritional delivery to the child’s evolving metabolic capacity.⁸

The stabilisation phase (Weeks 1–2) prioritises the correction of life-threatening metabolic derangements - hypoglycaemia, hypothermia, dehydration, and electrolyte imbalance - before attempting nutritional rehabilitation. F-75, a therapeutic milk formula providing 75 kcal and 0.9 g protein per 100 ml, is used exclusively during this phase. Its deliberately low protein and energy density reflects the risk of refeeding syndrome and “catch-up growth” overload in an acutely decompensated child; introducing higher-energy feeds too rapidly can precipitate cardiac failure, particularly in children with kwashiorkor-associated myocardial atrophy. Broad-spectrum antibiotic treatment is given to all SAM admissions regardless of visible infection, reflecting the high prevalence of subclinical sepsis in this population.

The rehabilitation phase (Weeks 2–6) transitions the child to F-100, a milk formula providing 100 kcal and 2.9 g protein per 100 ml, to promote rapid catch-up growth. Breastfeeding is continued throughout where applicable. Micronutrient supplementation - including zinc, copper, folic acid, multivitamin preparations, and potassium - is maintained. In children with kwashiorkor, the resolution of oedema is tracked daily: diuresis precedes weight gain, and caregivers and staff must be informed that apparent initial weight loss during the first days of treatment reflects fluid mobilisation rather than treatment failure.

Follow-up and sensory stimulation constitute a third dimension of care within the inpatient protocol: structured play and emotional stimulation, beginning in the rehabilitation phase, reduce the neurodevelopmental sequelae of prolonged severe undernutrition.

Community-Based Management and RUTF

The landmark work of Collins et al. (2006), published in The Lancet, established the evidence base for community-based management of acute malnutrition (CMAM), demonstrating that uncomplicated SAM could be managed safely and effectively in outpatient settings using ready-to-use therapeutic food (RUTF).⁹ RUTF - most commonly a peanut-based paste providing 500 kcal per 92 g sachet, enriched with vitamins and minerals - enabled treatment at scale without hospitalisation, transforming the operational reach of SAM management in settings where inpatient bed capacity was chronically insufficient.

CMAM protocols reserve inpatient care for children with complications: anorexia (failure to pass the appetite test), bilateral pitting oedema of Grade 3, or concurrent medical illness. Children who pass the appetite test and lack medical complications are enrolled in outpatient therapeutic programmes (OTP), attending weekly or fortnightly for MUAC measurement, weight, oedema assessment, clinical review, and RUTF distribution. Cure rates in well-functioning CMAM programmes consistently exceed 75%, with mean lengths of stay between six and ten weeks, compared to the 85–90% case fatality rates observed in the pre-CMAM era for SAM managed solely within overwhelmed inpatient facilities.

The clinical implications for syndrome differentiation in community settings are significant: children with kwashiorkor, particularly those with Grade 2 or 3 oedema, are at higher risk of deterioration in outpatient programmes and require more frequent review and a lower threshold for inpatient referral than their marasmic counterparts.

Syndrome-Specific Therapeutic Considerations

Children with kwashiorkor require particular attention to electrolyte management: hypokalaemia, hyponatraemia, and hypomagnesaemia are common and potentially fatal, and the routine addition of ReSoMal (rehydration solution for malnourished children) rather than standard ORS is recommended when rehydration is needed, given that standard ORS carries a sodium load that can precipitate fluid overload in oedematous children. Conversely, RUTF palatability presents a greater clinical challenge in kwashiorkor due to anorexia - nasogastric feeding may be necessary for a subset of children who cannot be motivated to eat voluntarily.

Children with marasmus, by contrast, are typically motivated to eat but require close monitoring for refeeding hypophosphataemia in the early rehabilitation phase if nutritional repletion is accelerated. Their superior alertness and preserved hunger drive contribute to faster OTP engagement and shorter treatment durations on average.

For a broader review of micronutrient supplementation strategies in the management of these children and their longer-term nutritional recovery, see our article on the role of micronutrient interventions .

Limitations and Methodological Considerations

Several important caveats constrain the interpretation of the clinical and epidemiological literature reviewed here.

First, the diagnostic criteria for kwashiorkor and marasmus have not been uniformly applied across the studies contributing to prevalence estimates and case fatality data. Some older surveys used the Wellcome classification, which defined kwashiorkor and marasmus partly on the basis of percent of expected weight-for-age - a criterion now largely superseded by WHZ and MUAC-based classifications. Comparisons across time periods or between sites using different classification systems require caution.

Second, the dichotomy between kwashiorkor and marasmus may be clinically misleading in populations with high mixed-presentation rates. The continuum model of SAM - recognising that the two syndromes represent different adaptive responses to similar underlying exposures rather than categorically distinct diseases - has not fully penetrated clinical training curricula, and facilities that apply different management tracks rigidly by diagnostic category may mismanage children who transition between presentations during treatment.

Third, community-based prevalence data from SSA are subject to seasonal variation that is rarely captured in single-round surveys. Acute malnutrition rates in the Sahel, for instance, peak in the lean season between June and September - often months before the harvest that will restore dietary adequacy - meaning that surveys conducted in other seasons systematically underestimate peak burden. Studies based on HDSS platforms that track individuals longitudinally offer a partial remedy, but are geographically limited.

Fourth, the relationship between kwashiorkor, aflatoxin exposure, and oxidative stress - one of the most plausible mechanistic frameworks for explaining the differential distribution of the two syndromes - remains incompletely characterised at the population level, with most evidence derived from case-control studies susceptible to residual confounding. Prospective cohort studies with serial biomarker measurement are needed to resolve these questions.

Finally, treatment outcome data from CMAM programmes vary substantially in quality across reporting sites. Programme data submitted to national nutrition information systems frequently under-report defaulter rates and over-report cure rates relative to independent evaluations. Pooled effectiveness estimates should therefore be interpreted as upper bounds rather than reliable point estimates for all implementation contexts.

Frequently Asked Questions

What is the single most important clinical sign distinguishing kwashiorkor from marasmus?

Bilateral pitting oedema is the defining diagnostic criterion of kwashiorkor and is absent in uncomplicated marasmus. A child with bilateral pitting oedema is classified as having SAM in the kwashiorkor category regardless of weight or MUAC, whereas a child with severe wasting (WHZ < −3 or MUAC < 115 mm) but no oedema is classified as having marasmus. In clinical examination, the oedema test - firm bilateral pressure on the dorsum of both feet for three seconds - should be performed on every child screened for malnutrition.

Why are children with kwashiorkor at higher risk of death than those with marasmus?

Children with kwashiorkor present with a broader and more severe range of metabolic derangements: hypoalbuminaemia impairs oncotic pressure and fluid balance, hepatic dysfunction disrupts clotting factor and immune protein synthesis, and dermatosis breaches the skin’s primary barrier against infection. The combination of immunosuppression, fluid dysregulation, and anorexia makes the clinical management of kwashiorkor substantially more complex, and the risk window for life-threatening complications - particularly septicaemia, hypoglycaemia, and electrolyte crisis - is both wider and less predictable than in marasmus.

Can a child have both kwashiorkor and marasmus simultaneously?

Yes. The category of marasmic-kwashiorkor applies to children who present with both severe wasting (WHZ < −3) and bilateral pitting oedema. This combined presentation carries the highest case-fatality risk of the three SAM categories. It is managed similarly to kwashiorkor - with the phase approach, F-75 in stabilisation, and careful electrolyte management - but requires heightened vigilance for rapid deterioration given the concurrent severity of tissue depletion.

What is RUTF and why is it used instead of hospital milk feeds in the community setting?

Ready-to-use therapeutic food (RUTF) is a lipid-based, nutrient-dense paste - typically peanut-based - that provides the protein, energy, and micronutrient content needed for rapid catch-up growth in children with uncomplicated SAM. Its key operational advantage is that it requires no water preparation, has a long shelf life at ambient temperatures, and can be administered at home by a caregiver following a brief demonstration. This makes it the cornerstone of outpatient therapeutic programmes where inpatient care is unavailable or impractical. Collins et al. (2006) established that RUTF-based community management achieves recovery rates equivalent to those of hospital-based F-100 rehabilitation, at substantially lower cost and with considerably higher coverage.⁹