Document Detail

Adequacy of human milk viscosity to respond to infants with dysphagia: experimental study.
Jump to Full Text
MedLine Citation:
PMID:  22230987     Owner:  NLM     Status:  MEDLINE    
OBJECTIVES: Determine viscosity variations of untreated human and pasteurized milk without and with thickening to adapt the diet to the needs of dysphagic infants hospitalized in the Neonatal Intensive Cara Unit (NICU).
MATERIAL AND METHODS: The authors altered the viscosity of natural infant powdered milk and, after thickening, determined and adopted a thickening standard for human milk. Untreated human and pasteurized milk was thickened in concentrations of 2%, 3%, 5% and 7% and the viscosity were determined every 20 minutes for a period of 60 minutes at a temperature of 37ºC.
RESULTS: The infant lactose formula thickened at concentrations of 2% and 3% produced viscosities of 8.97cP and 27.73 cP, respectively. The increases were significantly different after 1 hour. Inversely, untreated human milk at 2%, 3%, 5% and 7% produced diminished viscosity over time; the changes were more accentuated in the first 20 minutes. In pasteurized human milk, the 2% concentration had no variation in viscosity, but with the 3%, 5% and 7% concentrations, there was a significant decrease in the first 20 minutes with stability observed in the subsequent times.
CONCLUSION: In powdered milk, the viscosity increases over time; the viscosity in human milk diminishes. The results point out the importance not only of considering the concentration of the thickener but also the time being administered after its addition to effectively treat dysphagic infants.
Mariangela Bartha de Mattos de Almeida; João Aprígio Guerra de Almeida; Maria Elisabeth Lopes Moreira; Franz Reis Novak
Related Documents :
410957 - Organochlorine poisoning of ring-billed gulls in southern ontario.
21761407 - The development of the uncanny valley in infants.
1383727 - The load of genetic and partially genetic disease in man. iv. severe visual handicaps a...
21718117 - Costs of necrotizing enterocolitis and cost-effectiveness of exclusively human milk-bas...
17403827 - Breastfeeding and hospitalization for diarrheal and respiratory infection in the united...
24526007 - Variability of resistive indices in the anterior cerebral artery during fontanel compre...
Publication Detail:
Type:  Journal Article; Research Support, Non-U.S. Gov't    
Journal Detail:
Title:  Journal of applied oral science : revista FOB     Volume:  19     ISSN:  1678-7765     ISO Abbreviation:  J Appl Oral Sci     Publication Date:    2011 Nov-Dec
Date Detail:
Created Date:  2012-01-10     Completed Date:  2012-06-12     Revised Date:  2014-07-30    
Medline Journal Info:
Nlm Unique ID:  101189774     Medline TA:  J Appl Oral Sci     Country:  Brazil    
Other Details:
Languages:  eng     Pagination:  554-9     Citation Subset:  D; IM    
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Deglutition Disorders* / physiopathology
Food Additives / chemistry*
Infant Nutritional Physiological Phenomena / physiology*
Infant, Newborn
Intensive Care Units, Neonatal
Milk, Human / chemistry*
Time Factors
Reg. No./Substance:
0/Food Additives

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Full Text
Journal Information
Journal ID (nlm-ta): J Appl Oral Sci
Journal ID (iso-abbrev): J Appl Oral Sci
Journal ID (publisher-id): J. Appl. Oral. Sci.
ISSN: 1678-7757
ISSN: 1678-7765
Publisher: Faculdade de Odontologia de Bauru da Universidade de São Paulo
Article Information
Download PDF

Received Day: 23 Month: 7 Year: 2010
Accepted Day: 16 Month: 8 Year: 2011
Print publication date: Season: Nov-Dec Year: 2011
Volume: 19 Issue: 6
First Page: 554 Last Page: 559
PubMed Id: 22230987
ID: 3973454
DOI: 10.1590/S1678-77572011000600003

Adequacy of human milk viscosity to respond to infants with dysphagia: experimental study
Mariangela Bartha de Mattos de ALMEIDA1
João Aprígio Guerra de ALMEIDA2
Maria Elisabeth Lopes MOREIRA3
Franz Reis NOVAK4
1 MSc, Child and Woman Health, Speech-Language Therapist, Fernandes Figueira Institute, Oswaldo Cruz Foundation - FIOCRUZ, Rio de Janeiro, RJ, Brazil.
2 PhD in Public Health, Coordinator of Ibero American Program of Human Milk, Fernandes Figueira Institute, Oswaldo Cruz Foundation - FIOCRUZ, Rio de Janeiro, RJ, Brazil.
3 PhD in Public Health, Department of Neonatology, Fernandes Figueira Institute, Oswaldo Cruz Foundation - FIOCRUZ, Rio de Janeiro, RJ, Brazil.
4 PhD in Microbiology, Coordinator of The National Reference Center of a Human Milk Bank, Fernandes Figueira Institute, Oswaldo Cruz Foundation - FIOCRUZ, Rio de Janeiro, RJ, Brazil.
Correspondence: Corresponding address: Mariangela Bartha de Mattos de Almeida - Instituto Fernandes Figueira - Setor de Fonoaudiologia Hospitalar - Av. Rui Barbosa, 716 - Flamengo - Rio de Janeiro - Brasil - 22250-020 - Phone: 55 21 8852-3952 -


Neonatal nutrition is an important subject in community health and social planning policies, especially for questions arising from the lack of maternal nursing. Particularly in preterm newborns, nutrition takes on a predominant role for the child's overall development. Clinical and epidemiological data reinforce the conclusion that lack of human milk in this phase of life increases the incidence of diarrhea, respiratory infections, urinary infections, otitis media and necrotizing enterocolitis7 as well as depriving infants of components that contribute to adequate development.

Health professionals confront challenges in adequately feeding newborns13. In the Neonatal Intensive Care Unit (NICU), in general, preterm newborns present respiratory problems, asphyxia, cardiomyopathies, metabolic disturbances, genetic syndromes and congenital malformations, among others, that are often challenges for adequate nutrition.

Various obstacles can prevent adequate nutrition for this population during the hospitalization period: difficulties in feeding orally due to a delay in maturation and in acquiring the necessary oral sensory motor skills, prolonged use of tracheal-oral tubes and gastric probes, immaturity of the cardiorespiratory system, central nervous system and oro-facial musculature. These factors can also interfere with the nutritional process provoked by the disorganization of the normal pattern of sucking, hyper-sensibility to peri-touch and/or intra-oral stimuli. There also can be a lack of coordination in sucking, respiration and swallowing, resulting in complications such as food being aspirated into air passages when swallowing does not occur effectively6,11,14,15.

Moreover, babies that present with uncoordinated swallowing often do not have the necessary oral skills to suck the mother's breast and will require implementation of practices to adapt feeding practices, including a change in the consistency of the milk offered.

Souza, et al.16 (2001) confirmed that the thin liquid causes a lack of coordination among sucking, breathing and swallowing in patients. Thus, it is necessary to understand the importance of the consistency to create diets since consistency will influence the acceptance of and swallowing of food. In the National Dysphagia Diet2, seven rheological properties in food of major significance in the treatment of dysphagias have been identified. Among these, viscosity is the most important property for swallowing.

The goal of this study is to determine the variation in viscosity of untreated human and pasteurized milk without and with different degrees of thickening. This will offer health professionals information that can enable them to respond to the needs of the dysphagic infants hospitalized in the NICU without interrupting the use of human milk.


The study was undertaken in the Quality Control Laboratory of the Human Milk Bank of the Fernandes Figueira Institute - Oswaldo Cruz Foundation (IFF/FIOCRUZ). The study was approved by the IFF/FIOCRUZ's Committee for Ethics in Research on Human Beings. All human milk samples came from donors registered at the BLH-IFF, complying with legislation that regulates the establishment and functioning of the Human Milk Bank throughout the National Territory5.

An experimental study was undertaken in casual groups8 determined using version 13.0 of the SPSS program. The data collected were submitted to descriptive and statistical analysis using the Student's T test in variables with normal distribution and the Kruskal-Wallis test for non-parameter variables with a level of significance of 95% in accordance with the effects to be tested in each of six stages, described as follows:

1) Standardize the technique to determine viscosity of pumped human milk using Ford Viscometer cups

This step was necessary because although the Ford Viscometer cup has been used for various Newtonian fluids, it was not possible to identify prior publications that established the analytic technique necessary to be followed for determining the viscosity of human milk according to this method. It was necessary to determine which orifice of the Viscometer was most appropriate for the viscosity of fluid human milk and for thickened milk.

For the rheological characterization, we used the determination of cinematic viscosity according to standard NBR 58493 for the investigation of Newtonian fluids (such as milk) that extend the space of time for conduction between 20 s to 100 s for orifices 2, 3 and 4. Viscosity was determined by the Quimis brand model Q 280 Ford Viscometer cup.

Viscosity was determined by the flow time of the fluid and by gravity through the Viscometer. The time in seconds was converted into cinematic viscosity (centistoke) and then into dynamic viscosity (centipoise) using formulas based on the diameter of the Ford cup orifice.

2) Study the effect of time on the viscosity of untreated human milk pumped and pasteurized milk was maintained at a temperature of 37ºC with variation of ±2ºC

This study corresponds to two trials. Determination took place every 3 hours for a period of 9 hours. Repeated observations of the same product were aimed at controlling the temperature and reducing experimental errors. The temperature is a determinant of the product's quality and it was affected according its use in the service routine.

1.8 L of pumped untreated human milk was used with three repetitions of 600 mL to make triplicates with the volume of 200 mL in each trial. The repetitions were done with different samples of human milk and the triplicates composed by the same product. Each trial was undertaken separately and the determination of the viscosity was developed and is described in the results. Human milk was thawed; afterwards, with the help of a beaker, re-pouring the product created three repetitions with the 600 mL volume forming triplicates with volumes of 200 mL each. The pasteurization of human milk occurred as described by the ANVISA4 (2008). After thermal equilibrium at the study's temperature, the viscosity was determined.

To determine the kinematic viscosity in centistokes at this stage, orifice 2 was used and the following formula was used, provided by NBR 5849: Viscosity=[(2,388xt) - (0,007xt2) - 57,008]

Variable t represents the time of rapid flowing in seconds. The result was converted into centipoise.

3) Study the effect of time on the infant's powdered milk viscosity appropriate for the first semester maintained at 37ºC with a variation of ±2ºC.

In this study, we used six repetitions with 150 mL and analyzed changes in the viscosity at 20-minute intervals for a period of 60 minutes, maintaining the product in thermal equilibrium at 37ºC. We adopted first semester infant powdered milk purchased locally as a thickener; the concentrations were 2% and 3%. The viscosity changes were initially undertaken without the thickener and prepared according to manufacturer recommendations. This permitted recognition of the pattern of the viscosity. The same viscosity formula was used as that in the previous step.

4) Study the effect of time on the viscosity of infant powdered milk appropriate for the first semester, thickened at 2% and 3% and maintained at a temperature of 37ºC with a variation of ±2ºC.

We used 12 repetitions of the powdered milk. Six were thickened with the 2% concentration and six with the 3% concentration. The viscosity of each repetition was determined in four intervals of 20 minutes for a period of 60 minutes maintaining the product at thermal equilibrium at 37ºC. Thicken Up® (Nestlé Resource, São Paulo, Brazil) was the commercial thickener chosen, and it was purchased locally.

Considering the benefits from thickening, we based the testing on our clinical practice and used a thin consistency (1-50 cP) that enabled suction of the feeding bottle or feeding with the little cup. We followed the table recommended by the American Dietetic Association2.

Each concentration of the thickener was weighed on the Marte brand model A500 digital scale and afterwards the powdered milk was added according to the manufacturer's recommendation. The timing included a two-minute pause for thickening to be completed.

For the 2% concentration, the same formula was used as that of the previous study and for the 3% concentration, the viscosity was determined with orifice 3 and the following formula was applied:


5) Study the effect of time on the viscosity of untreated human milk, thickened in 2%, 3%, 5% and 7% concentrations and maintained at a temperature of 37ºC with a variation of ±2ºC.

For this study, a pool of human milk was created for testing the 2% and 3% concentrations and another pool for the 5% and 7% concentrations. Six duplicates were collected for pasteurization and six were untreated for each concentration. In each concentration, each duplicate was accompanied by a period of an hour and its viscosity was determined in 20-minute intervals for a total of four measurements.

For the 2%, 3% and 5% concentrations, the formula was applied for orifice 2. For the 7% concentration, viscosity was determined with orifice 3, and the corresponding formula was applied.

In studying 5% and 7% concentrations, the goal was to reach the necessary degree of thickening for infants with dysphagia, comparing it with the viscosity obtained in the concentration of infant powdered milk already known and used clinically at 2% and 3% concentrations.

6) Study the effect of time on the viscosity of pumped pasteurized human milk, thickened in 2%, 3%, 5% and 7% concentrations and maintained at a temperature of 37ºC with a variation of ±2ºC.

The study design followed the same model of the untreated pumped human milk thickened in the same concentrations.


Samples of untreated human and pasteurized milk maintained at 37ºC for a period of nine hours did not alter the viscosity values when compared by analysis of the variance to determine a 5% level of probability and a median value of 2.84 cP in untreated milk and 3.40 cP in pasteurized milk as observed in Table 1.

The viscosity determination values for first semester infant powdered milk was undertaken every 20 minutes after its preparation and maintained at 37ºC for a period of up to one hour, is presented in Table 2.

A 17.39% (p<0.05) reduction was observed over the first 20 minutes after preparation and was associated with the product's chemical stabilization. That is, in the initial period, the hydration of formula components takes place after its dissolution. It is worth pointing out that even though the product instantaneously dissolves, the subsequent hydration requires time to occur. This conclusion is reinforced by the fact that the viscosity is maintained stable with values that are statistically equal over 40 subsequent minutes.

The median value of viscosity of the powered milk was 4.49 cP. When compared with untreated pumped human milk or pasteurized pumped human milk at 37ºC, the powdered milk had a higher viscosity.

The addition of 2% of thickener in the powdered milk at 37ºC produced double the increase in viscosity compared to results from the powdered milk without added thickener during the first 20 minutes. The addition of 3% concentration increases the powdered milk's viscosity six-fold in relation to time zero.

Comparing all concentrations of thickener of raw pumped human milk analyzed, the average viscosity values decreased significantly in relation to time (Table 3).

In all analyzed concentrations, the decrease in the viscosity is more accentuated in the first 20 minutes, maintaining a stability for between 40 and 60 minutes, except in the 3% concentration, which continues to exhibit diminishing viscosity with the passage of time (p<0.05).

In relation to untreated pumped human milk without thickener, we observed that the addition of 7% of the thickener increased the viscosity up to nine-fold.

The addition of 2% of thickener in pasteurized pumped human milk did not modify its viscosity significantly over time. With the 3%, 5% and 7% concentrations, we verified that there was a significant decrease in the viscosity over time, mainly in the first 20 minutes. The tendency was to stabilize viscosity between 20 and 40 minutes (Table 4).

In relation to pasteurized pumped human milk without thickener, we verified that the addition of 7% of thickener increased up to eight-fold the viscosity of pasteurized pumped human milk observed at time zero.


The samples maintained at 37ºC for a period of nine hours of untreated human and pasteurized milk did not have altered viscosity values.

In powdered milk, the viscosity was maintained stable with statistically equal values over 40 subsequent minutes. In the powdered milk thickened at 2% and 3%, a considerable increase in viscosity over time was observed.

Compared with powdered milk, human milk produced an inverse viscosity behavior: the viscosity diminished with time. With the powdered milk, there was a progressive and rapid increase with a tendency to stabilize. In this way, we imagine that the immediate offer of human milk after the addition of thickener could minimize such effects and preserve the desired degree of thickening.

The results reinforce that the use of thickeners should take into consideration not only the concentration of the product as well as the time elapsing after preparation before administration is important in responding to the needs of the dysphagic infant.

From the theoretical point of view, the reduction in the viscosity of thickened human milk is associated with the presence of the amylase enzyme in human milk that promotes of partial hydrolysis of the starch contained in the thickener10. This leads to the reduction, proving that pasteurization does not cause total inactivity of this enzyme in human milk.

Therefore, amylase becomes an important ally in the nutrition of the dysphagic infants. This enzyme, in very small amounts in the first three or four months of life in the digestive tube of babies, is present in human milk9,12. Its presence in the thickened human milk facilitates the digestive process. This doesn't happen with the thickened infant powdered milk, as it will depend on the babies' enzyme. So, the thickening of the human milk is beneficial, as this technique maintains the human milk its exclusive nutritional quality and acts as a facilitator of the digestive process.

For the standardization of the thickening procedure of diet in the routine of hospital nutrition and speech therapy services, pasteurized pumped human milk with thickener could be administered in four consistencies: 2%, 3%, 5% and 7% according to the speech pathologist's evaluation.

The study points to the fact that the result of the thickening process for human milk is different than that with infant powdered milk. This could reflect the influence of amylase's action especially in untreated human milk leading to a decrease in the thickened product's viscosity.

A limiting factor in this study was the method for determining the viscosity. It requires large volumes of milk, and it is difficult to control the temperature.

Our results indicate that it is possible, in practical terms, to seek alternatives for thickening human milk with human milk components. The post-pasteurization viscosity variations arising from alterations in the quaternary structure of the milk proteins suggest that the paths for developing thickeners using the cream of human milk or with the help of bacteria should be investigated1. The thickeners could be able to promote the sweet coagulation of human milk, increasing its viscosity without acidifying it. It is worth noting that the food industry has used a similar technology for a long time to produce "flan" and skimmed creamy cheese. Studies of this type will contribute to new strategies for the feeding of the dysphagic infant, permitting the infant to benefit from the use of human milk.


The authors wish to thank the Foundation for the Support of Research of the State of Rio de Janeiro (FAPERJ) - Brazil for the financial support (Process: E- 26/102.873/2008).

1. Almeida JAG. Amamentação: um híbrido natureza-culturaRio de JaneiroFIOCRUZYear: 1999
2. American Dietetic Association (ADA)The national dysphagia diet: standardization for optimal careChicagoThe AssociationYear: 2002
3. Associação Brasileira de Normas Técnicas (ABNT)NBR 5849: Tintas - Determinação da viscosidade pelo copo FordRio de JaneiroABNTYear: 1986
4. BrasilAgência Nacional de Vigilância Sanitária (ANVISA)Banco de leite humano: funcionamento, prevenção e controle de riscosBrasíliaANVISAYear: 2008
5. BrasilAgência Nacional de Vigilância Sanitária (ANVISA)Resolução RDC nº 171 de 04 set. 2006. Dispõe sobre o regulamento, a implantação e o funcionamento de Bancos de Leite Humano no território nacionalDiário Oficial da UniãoBrasíliaDay: 05 Season: Set. Year: 2006 Seção 1. 33
6. Brock R. Recém-nascido prematuro, baixo peso e retardo do crescimento intra-uterinoBasseto MCA,Brock R,Wajnsztejn R. Neonatologia: um convite à atuação fonoaudiológicaSão PauloEditora LoviseYear: 19986773
7. Camelo JS Jr,Martinez FE. Lactoengenharia do leite humanoPereira GR,Leone CR,Alves N Filho,Trindade O Filho. Nutrição do recém-nascido pré-termoRio de JaneiroMedbookYear: 20081129
8. Gomes FP. Curso de estatística experimentalSão PauloEditora NobelYear: 1978
9. Martinez FE,Camelo JS Jr. Nutrition of the preterm infantsJ. Pediatr (Rio J)Year: 200177Suppl. 1S32S4014676891
10. Martins EC,Krebs VL. Effects of the use of fortified raw maternal milk on very low weight infantsJ Pediatr (Rio J)Year: 200985215716219319449
11. Mercado-Deane MG,Burton EM,Harlow SA,Glover AS,Deane DA,Guill MF,et al. Swallowing dysfunction in infants less than 1 year of agePediatr RadiolYear: 200131642342811436889
12. Moreira AR. Diversificação alimentar: o início da aprendizagem para uma vida saudável [Dissertação de Licenciatura]PortoFaculdade de Ciências da Nutrição e Alimentação, Universidade de PortoYear: 2005
13. Moreira MEL,Goldani MZ. Child is the father of man: new challenges for child healthCiênc Saúde ColetivaYear: 2010152321327
14. Neiva FCB,Leone CR. Sucking in preterm newborns and the sucking stimulationPro FonoYear: 200618214115016927619
15. Newman LA,Keckley C,Petersen MC,Hamner A. Swallowing function and medical diagnoses in infants suspected of dysphagiaPediatricsYear: 20011086
16. Souza BB,Martins C,Campos DJ,Balsini ID,Meyer LR. Nutrição e disfagia: guia para profissionaisCuritibaMethaYear: 2003

[TableWrap ID: t01] Table 1 

Effect of time on the viscosity of untreated human milk and pasteurized milk maintained at a temperature of 37ºC

Time (hours) Viscosity (cP) UHMP Viscosity (cP) PHMP
0 2.84±0.30a 3.40±0.67a
3 2.73±0.51a 3.59±0.84a
6 2.48±0.74a 3.40±0.74a
9 3.35±0.50a 3.98±1.05a

Values followed by same letter do not differ at 5% probability (p<0.05)

UHMP=Untreated Human Milk Pumped; PHMP=Pasteurized Human Milk Pumped

[TableWrap ID: t02] Table 2 

Effect of time on the viscosity of infant powdered milk maintained at 37ºC

Time (minutes) Viscosity (cP) Viscosity (cP) 2% Viscosity (cP) 3%
0 4.49±0.54a 8.97±1.11a 27.73±1.88e
20 3.71±0.49b 13.43±0.55b 47.74±1.40f
40 3.81±0.33b 15.58±0.54c 47.57±1.22f
60 4.08±0.47b 15.92±0.52d 48.07±1.26f

Values followed by same letter do not differ at 5% probability (p<0.05)

[TableWrap ID: t03] Table 3 

Effect of time on the viscosity of untreated human milk thickened maintained at 37ºC

Time (minutes) Viscosity (cP) 2% Viscosity (cP) 3% Viscosity (cP )5% Viscosity (cP) 7%
0 4.54±0.25a 5.05±0.48d 7.59±0.92h 25.55±0.37k
20 3.39±0.20b 3.58±0.32e 4.67±0.40i 20.77±0.25l
40 3.81±0.25c 4.25±0.27f 4.11±0.20j 20.50±0.28m
60 4.12±0.53c 3.65±0.28g 3.86±0.40j 20.68±0.20m

Values followed by same letter do not differ at 5% probability (p<0.05)

[TableWrap ID: t04] Table 4 

Effect of time on the viscosity of pasteurized human milk thickened maintained at 37ºC

Time (minutes) Viscosity (cP) 2% Viscosity (cP) 3% Viscosity (cP) 5% Viscosity (cP) 7%
0 4.06±0.82a 5.53±0.68b 7.98±0.87e 27.09±1.01h
20 3.58±0.64a 4.43±0.22c 4.82±0.35f 20.92±0.38i
40 3.57±0.34a 4.43±0.28c 4.80±0.18f 20.59±0.20i
60 3.60±0.31a 4.01±0.30d 4.57±0.16g 20.76±0.18i

Values followed by same letter do not differ at 5% probability (p<0.05)

Article Categories:
  • Original Articles

Keywords: Deglutition disorders, Dysphagia, Deglutition, Human milk, Viscosity, Rheology.

Previous Document:  Instruments to assess the oral language of children fitted with a cochlear implant: a systematic rev...
Next Document:  Development of a CD-ROM on written language for the continuing education of elementary school teache...