Evaluation of Relactation by the Supplemental Suckling Technique

By Odile Oberlin and Caroline Wilkinson, Action Contre la Faim (ACF)

Odile Oberlin is a paediatrician working in a Paris hospital and research institute. She works as a volunteer consultant with Action Contre la Faim on a regular basis in Afghanistan.

The authors would like to acknowledge the contributions of the Ministry of Public Health, Afghanistan, the ACF team in Afghanistan and Cecile Bizouerne, ACF Psychologist, to this article. Acknowledgements also to the Afghan Ministry of Public Health, the French Ministry of Foreign Affairs and UNICEF for proving the financial support for the nutrition programmes in Kabul.

This article describes ACFs experiences of managing malnourished infants under six months in an inpatient setting in a challenging environment, which leads them to raise key questions about managing this age-group.

Action Contre la Faim (ACF) supports the only structures treating severe malnutrition within three paediatric hospitals or wards in Kabul. The children treated in the nutrition centres are mainly less than 5 years old with infants less than six months comprising over one-third of the admissions (37.4% from January to May 2006). The problem of infant malnutrition in Kabul is not new. Field Exchange Issue 9, p16-17, highlighted the high mortality rates (17%) observed amongst infants less than 6 months old who had been admitted in the Therapeutic feeding Units (TFUs) in Kabul in 1999.

The 1999 WHO guidelines on the management of severe malnutrition¹ do not address the specific needs of infants under six months² and do not include breastfeeding support. Supportive care to reestablish breastfeeding is described in Integrated Management of Childhood Illness (IMCI) guidelines³. A chapter on the management of malnourished infants less than six months is included in resource materials⁴ collaboratively developed by the IFE Core Group⁵. ACF has developed their own protocols based on advice from experts, observation and evaluation of the responses of infants treated within therapeutic feeding programmes, the integration of recently developed materials regarding breastfeeding, and the experience gained through the psychosocial approach to treatment developed by ACF in recent years.

Table 1 Admission criteria
Type of beneficiaries	Admission
Infant below 6 months	. Difficulties breastfeeding (Mother Milk Insufficiency (MMI), baby too weak to suckle, cracked nipples.) and/or . Bilateral oedema and/or . Weight less than 2.5 kg for length less than 49 cm and/or . Infant is more than 6 months but less than 4 kg and/or . Infant has a W/H < 80% and infant not gaining or is losing weight
or
Infant above 6 months & < 4 kg

 

Admission criteria

In Kabul, many mothers complain of a lack of breastmilk and believe that this is due to stress and not eating enough good food. Mothers presenting with breastmilk insufficiency raise a number of challenges for treatment of infants in feeding centres. The admission criteria and treatment of these young infants in the TFUs in Kabul has evolved over time. In 2003, the criterion of 'mothers milk insufficiency (MMI)⁶, was added to the existing admission criteria. In June 2005, the criteria were amended further to admit infants with a weight-for-length less than or equal to 80% if the mother reported she was suffering from a 'lack' of breastmilk and the infant was not gaining or was losing weight at home. Admission criteria at the time of the study in 2006 for this age-group thus comprised a combination of anthropometric criteria, weight criteria and difficulties in breastfeeding (see Table 1).

The protocols that have been developed to manage this age-group involve increasing the production of breastmilk through the supplementary suckling technique (SST) (see Box 1). The aim is that infants under six months should be discharged when gaining weight and exclusively breastfed, independent of their anthropometric status. The SST for relactation has been implemented systematically in the Kabul Therapeutic Feeding Units (TFUs) since August 2004. This article describes one of a series of studies aimed at improving the management of these young infants and their mothers.

Study Objectives

The aim of the study was to evaluate the impact of the SST during the period of management in the TFU (including the support, advice and activities proposed by the psychosocial workers). Analysis was based on data found in registration books, therapeutic cards, psychosocial forms (included family history, medical, treatment, feeding history of the infant, caregiver-infant observations), MMI forms (which focus on observed and reported infant feeding practices and health status of mother) and interviews with mothers. The indicators assessed were weight gain (g/kg/d), increase in estimated breastmilk quantity during the period in the TFU⁷ and evaluation of SST application.

The study also planned to assess progress of the infants on discharge from the TFU, including weight gain, breastfeeding status, and feeding practice. Follow-up of the patients was based on analysis of the 'dry ration books' in which infants are registered at discharge and where their weights and heights are registered during follow-up.

The target group comprised infants under 6 months of age and infants aged 6 months or over with a weight of less than 4 kg.

Data for each of the infants were entered into a spreadsheet and analysed using excel. Anthropometric indexes were calculated using Epinut version 5 and were exported into the excel spreadsheet.

Limitations of the study

These included;

• The size of the study was limited to infants discharged between 02/01/06 and 23/04/06.
• Some therapeutic cards (14 %) were unavailable due to poor storage and/or retrieval systems.
• The data on the therapeutic forms were often incomplete, e.g. on breastfeeding or use of the SST.
• Weight and mid upper arm circumference (MUAC) measurements were typically reported as rounded up figures. This was despite weighing scales accurate to the nearest 10g being available in all of the TFUs.
• The proportion of infants that were lost to follow-up after discharge hindered outcome evaluation.

Age profile

Data were analysed on the management of 94 infants aged less than 6 months old and/or less than 4kg discharged between 2/01/06 to 23/04/06 from three TFUs - Ataturk TFU (n=25), Indira Gandhi TFU (n=36), and Maiwand TFU (n=33).

The mean age of the sample was 4 months (see Figure 1). There were more boys (63%) than girls (37%). Twin births accounted for 17 % (n=16) of these admissions. According to the recorded TFU data, 18 infants had some form of disability and/or developmental problem that could affect feeding (including cleft palates, and suspected Down's Syndromes).

Admission criteria

Given the lack of consensus on admission criteria for infants under 6 months, severe and moderate acute malnutrition are referred to according to the criteria commonly applied to infants and children from 6 - 59 months of age.

Thirty-five infants met the classic criteria for severe acute malnutrition in this age-group of whom:

• 8 had oedema
• 21 had W/H < 70 % with no oedema
• 6 infants were over 6 months and weighed less than 4 kg.

Nearly one-third (31%, n=29) of infants under six months were admitted due to MMI but were not severely malnourished. Of these, 26 had moderate acute malnutrition (70 % < W/H 80 %) and three infants were not malnourished according to anthropometric criteria⁸. Reasons for admission did not vary significantly according to gender or from one TFU to another.

Admission weights ranged from 1.2 to 4.4 kg, mean weight 2.6 kg (SD 0.6 kg). For infants <6 months old, the mean admission weight was 2.5 kg (SD 0.6 kg) and mean length⁹ was 50.7 cm (SD 4.6 cm)¹⁰.

Nearly one-third (32%, n=30) of the infants were less than 49 cm¹¹, of whom 24 infants weighed <2.5 kg and 3 infants weighed >=2.5kg (data on 3 infants were not available).

Table 2 Discharge according to admission criteria
	Total	Cured	Died	Default	Transfer	Criteria not met	Admission error
Severe malnutrition	29	20	2	3	1	1	2
Age ? 6 months and weight < 4 kg	6	3		1	1	1
Length < 49 cm and age < 6 months	27	13	2	2	4	6
Moderate malnutrition (70 % < W/H < 80 %) and Maternal Milk Insufficiency (MMI)	26	21	2	1	2
No malnutrition but MMI	3	3
MMI but anthropometric status unknown	3	1	1
Total	94	61 (64.9%)	7 (7.4%)	7 (7.4%)	8 (8.5%)	9 (9.6%)	2 (2.1%)

 

Anthropometric indices

As weight-for-length expressed in % of the median (W/H%M) using the National Centre for Health Statistics (NCHS) references cannot be calculated for infants with a length of < 49 cm, this index is only presented for infants >=49cm. Instead, weight-for-age z score (WAZ) and length-for-age z score (WHZ) were calculated for infants <49cm.

Infants less than 49 cm (n=30)

The mean age was 2.2 months and mean weight was 2.2 kg, ranging from 1.4 to 3.4 kg. In general, the weight of the infants <49cm increased with length (Figure 2) and the lengths of the infants generally increased with age (see Figure 3). However WAZ data in Figure 4 shows that the older infants tended to be more underweight. Figure 5 also shows a decline in HAZ with increasing age that suggests older infants are more chronically malnourished or stunted than the younger infants.

Infants >=49 cm (n=61)

The WAZ and HAZ indexes for infants ? 49 cm showed the same deteriorating trend with age as in infants <49cm. The W/H%M was also determined for this group (Figure 6) although there is no clear trend with age.

Outcomes

Table 2 gives the type of discharge according to criteria for admission. Seven children died during their stay in the TFUs (7.4 %, n=7).

Causes of death were septicaemia (2), pneumonia (1), aspiration - probably due to pulmonary infection (2), and unknown (2). Out of the 35 infants who were severely malnourished, 2 died (5.7%). This is lower than in other studies for this age-group where mortality rates as high as 19 % have been reported for infants in TFUs^12,13.

Discharge as 'cured'

Table 3 summarises the weight gains by admission criteria for infants discharged as 'cured'.
The TFUs discharge protocol is that infants under six months are discharged when they are gaining weight on breastmilk alone (i.e. exclusively breastfed). However, data on breastfeeding status on discharge (55/61) showed that only 29% (16/55)) of the infants discharged as 'cured' actually met the full criteria of 'discharge on breastmilk alone.'

The remainder (61% (37/55)) were receiving either full or partially diluted F100 milk (DF100) supplement on discharge. Discharge on supplemental milk differed by TFU, reflecting a clear-cut management difference between centres (see table 4).

Of the 26 infants admitted with MMI and with moderate acute malnutrition, only seven of the infants discharged as 'cured' were on breastmilk alone. One infant was discharged on half dose of diluted F100 and 13 infants were discharged on the full dose of diluted F100. Of the infants admitted with MMI who were not malnourished (n=3), none were discharged on breastmilk alone. One was discharged on half dose of DF100.

Failure to meet anthropometric discharge criteria

Infants who weighed < 2.9 kg at discharge or infants admitted on W/H%M criteria but had achieved only 80% W/H%M at discharge were classified as 'criteria not reached'. In reality, however, many of these infants were gaining weight and almost half (4/9) were discharged feeding on breastmilk alone. Their mean weight gain was 12.2g/kg/day (0.7-30.3 sd 10.4). Consequently, for the purposes of analysis, this sub-group of infants were grouped together with the infants discharged as cured. Overall, only 20 out of 64 infants who were discharged as 'cured' or for whom anthropometric discharge criteria were not met, were discharged on breastmilk alone.

Evaluation of supplemental suckling technique (SST)

Information on the use of SST was taken from the therapeutic cards. Of the 94 infants, 81 cards were retrieved and studied. Since breastfeeding support or the SST was not specifically recorded on the therapeutic cards, this analysis relied on inferring how the technique was used. This was based on the recorded reduction in the amounts of diluted F100 given to the infants, according to the protocol for the SST (see Box 1). A reduction in DF100 was recorded in 25/81 infants studied (30.9%). There was no indication from the cards as to how many (failed) attempts there were or what breastfeeding support was offered to the infants.

Where details of supplemental suckling were recorded (n=25), the method worked well in most of the infants. These showed full or partial attempts at stopping the supplementary milk, in accordance with the protocol. For 20 infants, good rates of weight gain during the rehabilitation phase were maintained after reducing diluted F100 (the mean weight gain was 11 g/kg/day during these two periods). This indicates that the mothers were supplying adequate quantities of breastmilk to cover the maintenance and growth needs of their infants. These 20 infants were discharged on breastmilk alone, and showed an estimated increase in breastmilk of 384 ml/day from admission in the TFU to discharge, with an average estimated daily output over the 5 days prior to discharge of 747 ml/day of breastmilk¹⁴.

In the TFUs where the SST was rarely performed, staff reported that this was because mothers were not happy when they saw that the quantity of therapeutic milk given to the infant was reduced and stopped.

Table 3 Weight gains for infants discharged as 'cured' by admission criteria
Admission criteria	Number 'cured'	Weight gain
Infants admitted with severe malnutrition (WH%M<70% and/or oedema)	20	17.1 g/kg/day (-1.8 - 43.8, sd 10.5)
? 6 months and < 4 kg	3	11.2g/kg/day (5.3 - 14.5, sd 5.1)
Infants <49cm	12*	21.5g/kg/day (3.5-50)
Infants with MMI and moderately malnourished	21	15.3g/kg/day (4.2-30, sd 6.6)
Infants with MMI but no malnutrition (WH ? 80%)	3	13.3g/kg/day (5.8-21.4, sd 7.8)
Total number of infants 'cured'	59

*Another infant was discharged at < 2.9 kg (2.71 kg) with a weight gain of 6.5g/kg/day, after 30 days in the centre.

 

Table 4 Milks infants were receiving upon discharge by TFU
	Ataturk	Maiwand	Indira Gandhi	Total
Discharged on diluted F100 (full or partial dose)	3 (=)	14 (*=)	20	37
Discharged on breastfeeding alone	9	6	1	16
Unknown	0	0	2	2
Total	12	20	23	55

* Two infants had stopped the DF100 altogether but then returned to the milk
= One infant's intake had been reduced to a half dose of DF100

 

Mothers therefore would want to go home. However this was not the experience of the TFUs where the SST was more successful. Also, of the seven infants who defaulted, the default does not appear to correspond to a decrease in the quantity of therapeutic milk given to the infants.

Where infants were not receiving therapeutic milk using the SST, the diluted F100 was given with a large cup and a spoon. This is a very 'passive' feeding technique for the infant and may cause aerophagia and feeling of satiety, thereby reducing appetite while increasing the risk of aspiration of the milk into the lungs. The use of a small cup allows the infant to drink actively, which is more appropriate where use of the SST is really not possible.

Staff in the TFUs did not show mothers how to bottle-feed or how to prepare artificial milks out of fear that this could encourage mother to adopt artificial feeding. Consequently, twothirds of mothers were discharged using supplemental milk with no instruction on how to safely prepare feeds at home.

Follow-up of infants on discharge

A total of 64 of the discharged infants met the criteria for follow-up¹⁵. Transfers, defaulters, and admission errors were not followed up. Of 64 infants, only 26 were available for follow up - 67% (10/15) of discharges were from Ataturk TFU, 58% (14/24) from Maiwand TFU and only 8% (2/25) from Indira Gandhi TFU. As Indira Ghandi is the largest paediatric hospital and of national repute, it is often the first port of call for the mothers from distant provinces. The follow-up of infants discharged from Indira Gandhi Hospital was also limited by the issuance of a 'dry ration identification number' at discharge that was not linked to their TFU admission number. The follow-up of children discharged from Ataturk and Maiwand was easier, due to the notes taken in the dry ration book and the possibility of attending the dry ration days at the TFU.

Of the 26 infants:

• Fourteen of 20 infants who were discharged on breastmilk alone were followed up. These infants had an average weight gain of 113g/week during the period of follow up (range -50g to + 300g).
• Twelve of 44 infants who had been discharged whilst still receiving a milk supplement were followed up. Average weight gain in this group was 120g/week (range minus 50g to +290g).

In both groups, one infant lost weight at a rate of 50g/week during the period of follow up and all of the others gained weight. Interpretation of these figures should be done with caution, given that two-thirds of infants who were receiving milk supplements on discharge were not followed up. There was also a large variation in both the number of weeks for which infants were recorded for follow up and the average weight gains per week between the infants in both groups.

Ten in-depth interviews were carried out with mothers who returned after discharge from the TFUs. Two reported a medical problem since discharge. Although all the mothers were still breastfeeding their infants, nine mothers were also giving their infants powdered milk and two were adding biscuits to the milk (infants aged 5.5 and 7 months). Powdered milk was either infant formula or powdered milk bought from the market taken from large bags (shir e kilogaki). The latter was much less expensive than the infant formula, but both the composition and the poor storage conditions of this powdered milk meant it was inappropriate for the needs of these young infants. On the basis of interviews with mothers, it was apparent that the way they reconstituted the powdered milk meant over-dilution (1 or 2 spoons for half a glass of water). Powdered milk was usually given after breastfeeding, but sometimes before breastfeeding. All the mothers reported using a cup and spoon to give the powdered milk.

Conclusions

The TFUs in Kabul show a much higher proportion of admissions of infants under 6 months than in TFUs in other countries. The treatment of these infants is more complex than that for older children. The high presentation of infants <49cm may represent low birth weight babies due to prenatal malnutrition or preterm birth who have failed to 'catch up', as well as infants who became malnourished after birth. There are no reliable data on the rate of low birth weight babies in Kabul.

The study suggests that the SST is a feasible strategy for the management of severe malnutrition within the Afghan hospital environment. Where SST is correctly applied, there are good results up until discharge. However, the SST was not well implemented overall, with many infants being discharged on mixed feeding. In the TFUs, these infants were still considered 'cured' and even where infants are discharged on breastmilk alone, it appears that mothers often reverted to mixed feeding at home.

We found that TFUs staff often considered the admission criteria vague and difficult to apply and were also confused by the treatment protocols and discharge criteria. In addition, it was counter-intuitive for staff who are used to older children remaining on a full supplement until discharge, to decrease daily quantities of milk for infants. Furthermore, these infants demanded a very different and more intense follow-up on the part of staff.

This study begs the question as to whether the admission criteria used in the Kabul context are the best ones, given the profile of infants identified for admission within the TFU. Mothers with lactation difficulties need lactation support, but the TFU may not be the place to provide this.

Inconclusive trials have been carried out within outpatient facilities in Kabul, employing a period of observation and support to women who complained of breastfeeding difficulties. The majority of the mothers in these trial interventions did not stay in the centre very long as they did not receive the supplementary milk they wanted.

There is a weak evidence base for the protocols and selection and discharge criteria for the identification and management of acute malnutrition in infants under six months. International guidance is unclear and sometimes contradictory. This poses significant challenges in deciding how best to treat these infants and in setting admission criteria - at which point do the benefits of the services and support offered from the therapeutic feeding unit in reducing morbidity and mortality outweigh the risks involved in maintaining infants in in-patient care and the introduction of artificial milk products? The absence of clear guidance for these infants often makes advocacy for inclusion of these young infants into care protocols more challenging when working within Ministry of Health structures or within national nutrition guideline frameworks.

Key policy and practical recommendations have emerged from this study for ACF programming, i.e. the need to prioritise improved care of young infants and in particular, the application of the SST. There is now an opportunity to develop the therapeutic protocols and tools for infants and include these in the new Afghan national protocol for the treatment of severe malnutrition. This opportunity must be grasped so that the needs of these young infants and their nursing mothers are adequately and appropriately catered for.

For further information, contact: Caroline Wilkinson, email: cwilkinson@actioncontrelafaim.org

---

¹Management of severe malnutrition: a manual for physicians and other senior health workers. Geneva, World Health Organisation,1999.

²Severe malnutrition: report of a consultation to review current literature. Geneva, World Health Organization, 6-7 September 2004. http://www.who.int/nutrition/publications/malnutrition/en/

³Supportive care to reestablish breastfeeding is described in IMCI Management of the child with a serious infection or severe malnutrition: guidelines for care at the first-referral level in developing countries (p99-104).

⁴Module 2 for health and nutrition workers in emergency situations. UNICEF, UNHCR, WHO, WFP, ENN, IBFAN, TdH and collaborators. Version 1.0. November 2004. Available at http://www.ennonline.net/ife/

⁵The IFE Core Group is an interagency collaboration concerned with policy guidance and capacity building on infant and young child feeding in emergencies. Current members are WHO, UNHCR, UNICEF, WFP, IBFAN-GIFA, CARE USA, ACF and ENN. See http://www.ennonline.net/ife

⁶Maternal Milk Insufficiency (MMI) is where a mother reports a lack of breastmilk, however there is no quantifiable measure of this on admission.

⁷This data is not presented in this field article but is available in the full study report from the authors (see contacts at the end).

⁸An additional 3 infants had missing length data (therefore no W/H calculated).

⁹Length data on 91/94 was available.

¹⁰The lengths of 3 children were not measured. These children weighed 1.5, 1.5 and 1.7 kg respectively (they were aged 1 day, 2 months and 1.5 months respectively).

¹¹One infant was classed in the category ? 6 months and <4 kg).

¹²M. Golden, Y. Grellety. The management of acute severe malnutrition, 2002

¹³Golden M. Comment on including infants in Nutrition Surveys: Experience of ACF in Kabul City. Field Exchange 2000; 9: 16 - 17.

¹⁴The mean breast milk output during the five first days after admission and the five last days before discharge for the 20 infants that left the centre exclusively breastfeeding. Assuming an infant required 110 kcal/kg/day for maintenance of their body weight and 5 kcal for each gram of weight gain, it was possible to estimate the daily energy intake taken for maintenance and weight gain from daily weights measured. It was assumed that the energy content of breast milk is 70 kcal per 100 ml. The breastmilk output was estimated, on this basis, assuming that it was the balance of energy consumed when the equivalent of energy from the supplemental milk was subtracted from the energy needs, according to the weight gain of the infant.

¹⁵Among the 70 infants who were discharged home from the TFUs, 6 infants were discharged after the 13th April 2006, which was too late for their consideration in follow-up for the purposes of this study.