Kangaroo mother care can improve the short-term outcomes of very preterm infants

Abstract

BACKGROUND:

Kangaroo mother care (KMC) is a cornerstone of preterm infant management. The purpose of this study was to estimate the effectiveness of daily prolonged KMC in very preterm infants and its influence on neonatal morbidity and short-term outcomes, and breastfeeding optimization.

METHODS:

Research included 52 very preterm infants. According to the KMC duration newborns were divided into two groups; Group1 of 22 infants (42.3%) – KMC lasted more than 3 hours/day, Group 2 of 30 infants (57.7%) – KMC lasted less than 3 hours/day.

RESULTS:

Nosocomial sepsis occurred less frequently in Group 1 versus Group 2 (OR = 10.50; 95% CI 1.23–89.67, p = 0.012). Incidences of BPD, NEC, IVH I–II grades, the duration of parenteral nutrition, and growth parameters have not been different between groups (p > 0.05). Breastfeeding rates at discharge prevailed in Group 1 (OR = 3.70; 95% CI 1.16–11.86, p = 0.025). The most important factors for nosocomial sepsis: combination of parenteral nutrition duration and daily prolonged KMC, as a preventing factor (p = 0.002). Combination of the neonatal intensive care unit (NICU) treatment duration, type of enteral feeding, and mother’s age add as factors that have important influence on breastfeeding prolongation (p = 0.009).

CONCLUSION:

Nosocomial infection prevention and breastfeeding optimization are profitable outcomes of daily prolonged KMC in very preterm infants. No significant differences in the BPD, NEC, IVH I–II grades incidences, duration of parenteral nutrition, and growth parameters were found between studied groups. Combination of long-lasting KMC and short-term parenteral nutritionis a significant factor for nosocomial sepsis prophylaxis.

Keywords

Very preterm infants kangaroo mother care breastfeeding nosocomial infection

1 Introduction

Kangaroo mother care (KMC) is a corner stone of preterm infant management that provides them with warmth, breastfeeding, safety, protection against infection, and hospital stay reduction [1 –4]. Kangaroo method includes skin-to-skin contact and breastfeeding that provides infantswith all the nutrients necessary for growth, development and health, early discharge from the hospital and, of course, the incredible power of parental love, care and tender embrace. There are two models of KMC practice: intermittent KMC for several hours per day and continuous KMC for 24 hours per day [3, 5].

To date, there is no consensus on the duration of Kangaroo mother carethat is necessary to optimize breastfeeding and its other profitable effects [1]. Therefore, the purpose of the present study was to estimate the effectiveness of prolonged Kangaroo mother care in very preterm infants and its influence on neonatal morbidity and short-term outcomes, and breastfeeding optimization.

2 Methods

2.1 Study design and participants

This retrospective research included 52 premature infants with gestational age 28/1–31/6 weeks which were born between March and September, 2016 at the Level 3 Children’s Hospital. Children were treated in the neonatal intensive care unit (NICU) where developmental careis implemented. All infants included in the study had KMCwith their mothers under the nurses’ supervision.

The influence of KMC duration on short-term outcomes was studied. The endpoints were as follows: late-onset sepsis (LOS), necrotizing enterocolitis (NEC), bronchopulmonary dysplasia (BPD), intraventricular hemorrhage I and II grade, artificial ventilation, parenteral nutrition duration, breastfeeding rate at discharge, growth parameters as weight at discharge in percentiles (evaluated with Olsen growth calculator), and daily weight gain.

According to the KMC duration, newborns were divided into two groups; group 1 of 22 infants (42.3%) – KMC lasted more than 3 hours/day, group 2 of 30 infants (57.7%) – KMC lasted less than 3 hours/day. The inclusion criteria for both groups was the duration of KMC per day, which was determined by mother’s possibility to spend time with baby and provide kangaroo care for her infant.

2.2 Statistical analysis

All computations were performed using Stat Soft STATISTICA Version 13 (Tulsa, OK). Quantitative data was presented as the median and inter quartile range (IQR; 25th to 75th percentiles). For qualitative parameters, absolute and relative frequencies were presented. The Mann-Whitney U-test was used to compare numerical data. Qualitative parameters were analysed by use of a 2×2 contingency table and Fisher’s exact test, Odds ratio (OR) and 95% confidence intervals (95% CI). Significance was assumed at p < 0.05.

Pseudo-randomization was performed with the Propensity score matching technique using the Kernel method and finding the nearest neighbor. Multiple logistic regression analysis was performed in order to analyze the factors that have an impact on the short-term outcomes of study population. The statistical significance of the obtained results was evaluated using Wald test. The probability of an event for some case was calculated by the formula $P = \frac{1}{1 + e^{- z}},$ where p is the probability of the predicted event; z = a + b₁^*X₁ + b₂^*X₂ + … b_n^*X_n; e – mathematic constant 2.72; a – model constant; X1, X2, Xn – values of independent variables; b1, b2, bn – coefficients, the calculation of which is the task of binary logistic regression; n – a serial number of the predictor included in the equation.

The building of the logistic regression model was carried out by the method of stepwise exclusion of prognostic factors with the definition of the minimum set of predictors by the estimation of Nagelkerke square. Results were considered statistically significant at p < 0.05.

3 Results

3.1 Patient’s characteristics

There were 17/22 (77.3%)boys and5/22(22.7%) girls in group 1 and 20/30 boys (66.7%) and 10/30 (33.3%) girls in group 2. The two groups were comparable for maternal clinical characteristics with the exception of maternal age, that was greater in Group 1 (p = 0.009), and newborns’ characteristics before KMC have been started (Table 1). To increase confidence in the results, pseudo-randomization was performed using the Propensity score matching technique.

Table 1
Characteristics of the study population

Group 1 Group 2

KMC more than 3 hours/day KMC less than 3 hours/day p value

(n = 22) (n = 30)

Maternal clinical characteristics

Maternal age (years), Me [Lq; Uq] 33.0 [30.0; 36.0] 29.0 [26.0; 33.0]^* 0.009

Gestational hypertension, preeclampsia/eclampsia, n (%) 7 (31.8) 12 (40.0) 0.544

Maternal infection, n (%) 8 (36.4) 9 (30.0) 0.632

Antenatal steroid, n (%) 20 (90.9) 26 (86.7) 0.639

Maternal hemorrhage, n (%) 3 (13.6) 6 (20.0) 0.547

Fetal hypoxia, n (%) 8 (36.4) 10 (33.3) 0.822

Characteristics of newborns

Gestation age, Me [Lq; Uq] 29.92 [29.14; 31.57] 30.43 [28.86; 31.28] 0.833

Birth weight (grams), Me [Lq; Uq] 1222.5 [1050.0; 1530.0] 1270.0 [1020.0; 1480.0] 0.920

Birth weight (percentile), Me [Lq;Uq] 30.5 [8.0; 53.0] 25.5 [7.0; 54.0] 0.949

Small for gestational age, n (%) 6 (27.3) 7 (23.3) 0.742

Apgar score at 1 min, Me [Lq; Uq] 9.0 [8.0; 10.0] 7.5 [5.0; 9.0] 0.057

Apgar 5 score at 5 min, Me [Lq; Uq] 10.0 [9.0;10.0] 10.0 [8.0; 10.0] 0.220

Primary resuscitation with ventilation, n (%) 11 (50.0) 9 (30.0) 0.143

Heart massage and epinephrine administration, n (%) 1 (4.5) 0 (0) 0.241

Respiratory distress syndrome (RDS), n (%) 10 (45.5) 16 (53.3) 0.578

Early-onset sepsis, n (%) 3 (13.6) 3 (10.0) 0.688

NICU treatment duration (days) 9.5 [7.0; 12.0] 7.0 [5.0; 15.0] 0.462

	Group 1	Group 2
Maternal clinical characteristics
Maternal age (years), Me [Lq; Uq]	33.0 [30.0; 36.0]	29.0 [26.0; 33.0]^*	0.009
Gestational hypertension, preeclampsia/eclampsia, n (%)	7 (31.8)	12 (40.0)	0.544
Maternal infection, n (%)	8 (36.4)	9 (30.0)	0.632
Antenatal steroid, n (%)	20 (90.9)	26 (86.7)	0.639
Maternal hemorrhage, n (%)	3 (13.6)	6 (20.0)	0.547
Fetal hypoxia, n (%)	8 (36.4)	10 (33.3)	0.822
Characteristics of newborns
Gestation age, Me [Lq; Uq]	29.92 [29.14; 31.57]	30.43 [28.86; 31.28]	0.833
Birth weight (grams), Me [Lq; Uq]	1222.5 [1050.0; 1530.0]	1270.0 [1020.0; 1480.0]	0.920
Birth weight (percentile), Me [Lq;Uq]	30.5 [8.0; 53.0]	25.5 [7.0; 54.0]	0.949
Small for gestational age, n (%)	6 (27.3)	7 (23.3)	0.742
Apgar score at 1 min, Me [Lq; Uq]	9.0 [8.0; 10.0]	7.5 [5.0; 9.0]	0.057
Apgar 5 score at 5 min, Me [Lq; Uq]	10.0 [9.0;10.0]	10.0 [8.0; 10.0]	0.220
Primary resuscitation with ventilation, n (%)	11 (50.0)	9 (30.0)	0.143
Heart massage and epinephrine administration, n (%)	1 (4.5)	0 (0)	0.241
Respiratory distress syndrome (RDS), n (%)	10 (45.5)	16 (53.3)	0.578
Early-onset sepsis, n (%)	3 (13.6)	3 (10.0)	0.688
NICU treatment duration (days)	9.5 [7.0; 12.0]	7.0 [5.0; 15.0]	0.462

^*– p value < 0.05.

3.2 Outcomes

Nosocomial sepsis occurred significantly less frequently in infants who had KMC more than 3 hours per day versus those, who had KMC less than 3 hours per day (1/22 vs 10/30; OR = 10.50; 95% CI 1.23–89.67, p = 0.012). Breastfeeding rates at discharge significantly prevailed in group 1 versus group 2(15/22 vs 11/30; OR = 3.70; 95% CI 1.16–11.86, p = 0.025). No significant differences in the BPD, NEC, IVH I–II grades incidences, duration of parenteral nutrition, growth parameters were found between studied groups (p > 0.05) (Table 2).

Table 2
Comparative clinical characteristics of newborns depending on the KMC duration

Group 1 Group 2 p Odds Ratio (95% CI)

KMC more than 3 hours KMC less than 3 hours/day

(n = 22) (n = 30)

n % N %

BPD 4 18.2 4 13.3 0.628 0.69 (0.15–3.14)

Nosocomial infection 1 4.5 10 33.3^* 0.012 10.5 (1.23–89.67)

NEC 0 0 2 6.7 0.220 –

Breastfeeding at discharge 15 68.2 11 36.7^* 0.025 3.70 (1.16–11.86)

Intraventricular hemorrhage I–II stages 5 22.7 8 26.7 0.742 1.24(0.34–4.46)

Duration of ventilation 2.0 [2.0; 5.0] 2.0 [1.0; 5.5] 0.981 –

Duration of parenteral nutrition (days) 12.0 [7.0; 15.0] 12.0 [7.0; 19.0] 0.747 –

Weight at discharge (percentille) 16.54 [6.0; 2.0] 11.0 [5.0; 24.0] 0.876 –

Weight at discharge ≤10th percentile 10 45.5 13 43.3 0.875 0.92 (0.30–2.78)

Weight at discharge ≤3th percentile 2 9.1 4 13.3 0.639 1.54 (0.26–9.26)

Daily weight gain (g) 21.75 [18.67; 25.35] 21.30 [18.10; 24.56] 0.761 –

	Group 1	Group 2	p	Odds Ratio (95% CI)
BPD	4	18.2	4	13.3	0.628	0.69 (0.15–3.14)
Nosocomial infection	1	4.5	10	33.3^*	0.012	10.5 (1.23–89.67)
NEC	0	0	2	6.7	0.220	–
Breastfeeding at discharge	15	68.2	11	36.7^*	0.025	3.70 (1.16–11.86)
Intraventricular hemorrhage I–II stages	5	22.7	8	26.7	0.742	1.24(0.34–4.46)
Duration of ventilation	2.0 [2.0; 5.0]		2.0 [1.0; 5.5]	0.981	–
Duration of parenteral nutrition (days)	12.0 [7.0; 15.0]		12.0 [7.0; 19.0]	0.747	–
Weight at discharge (percentille)	16.54 [6.0; 2.0]		11.0 [5.0; 24.0]	0.876	–
Weight at discharge ≤10th percentile	10	45.5	13	43.3	0.875	0.92 (0.30–2.78)
Weight at discharge ≤3th percentile	2	9.1	4	13.3	0.639	1.54 (0.26–9.26)
Daily weight gain (g)	21.75 [18.67; 25.35]		21.30 [18.10; 24.56]	0.761	–

^*–p value < 0.05; 95% CI –95% Confidence Intervals.

Multiple logistic regression analysis was performed in order to analyze the factors that have an impact on nosocomial sepsis occurrence. The presence of nosocomial sepsis (y = 1) and its absence (y = 0) were chosen as the binary dependent variables (y). Both quantitative (gestational age (GA), birth weight, NICU treatment duration, duration of ventilation, parenteral nutrition duration) and qualitative characteristics (gender of child) potentially affecting the nosocomial sepsis occurrence were chosen as independent variables. Long-lasting KMC more than 3 hours per day that potentially has positive influence in preventing nosocomial infections in preterm infants was also included in analysis as independent qualitative characteristic. Male gender was coded as “1”, female –“0”; KMC more than 3 hours per day was coded as “1”, shorter duration of KMC less than 3 hours - “0”.

Multiple logistic regression analysis showed factors which were the most important for nosocomial sepsis occurrence, in particular, it was a combination of parenteral nutrition duration (the longer the duration, the greater the likelihood of sepsis occurrence) and KMC more than 3 hours per day, as preventing factor (Table 3). Other factors that were studied didn’t show the significant indices.

Table 3

Statistical characteristics of the multiple logistic regression factors potentially able to have an impact on nosocomial sepsis occurrence in study population

Coefficient	Standard Error	Wald Statistic	P value
Constant	4.441	1.447	9.414	0.002
Parenteral nutrition duration	–0.244	0.0855	8.169	0.004
KMC (more than 3 hours per day)	2.542	1.248	4.152	0.042

According to the results of the analysis, multiple regression equation was created: $\begin{matrix} z = 4.441 - (0 . 244^{*} parenteral nutrition \\ duration) + (2 . 542^{*} KMC more than \\ 3 hours / day) . \end{matrix}$

3.3 Examples of calculating the probability of nosocomial sepsis in very preterm infants

Example 1. Male infant, GA 29 + 5, G1P2, birth weight 1310 g (46th percentile), cesarian section. Mother’s age –29 years. Apgar score 7/10. Duration of the NICU treatment –4 days, total hospital stay duration –48 days. Baby had RDS, received surfactant therapy, duration of ventilation –1 day, followed by the CPAP-therapy. Enteral feeding was started on day 0, parenteral nutritionlasted 26 days. During hospital stay infant had irregular KMC less than 3 hours/day.

Multiple regression equation was created for this infant: $\begin{matrix} z = 4.441 - (0 . 244^{*} duration of parenteral \\ nutrition) + (2 . 542^{*} KMC more than 3 hours / day) \\ z = 4.441 - (0 . 244^{*} 26) + (2 . 542^{*} 0); \\ z = - 1.9 \end{matrix}$

Now calculate the probability of nosocomial sepsis occurrence in this infant:

P = 1/(1 + e^-1.9) = 0.870. Thus, the probability of nosocomial sepsis is 0.870 or 87.0%. During hospital treatment this patient had nosocomial sepsis with clinical manifestation and bacteriologically proven data.

Example 2. Male infant, G3P2, GA 28 + 5, birth weight 1960 g (99th percentile), vaginal delivery. Mother’s age –33 years. Apgar score 1/5, baby had ABCD-steps of resuscitation. Duration of the NICU treatment –5 days, total hospital stay duration –56 days. Baby had RDS, received surfactant therapy, duration of ventilation –2 day, followed by the CPAP-therapy. Infant developed mild BPD. Enteral feeding was started on day 1, parenteral nutrition duration –22 days. During hospital stay infant had daily KMC with duration more than 3 hours/day.

Multiple regression equation for this infant: $\begin{matrix} z = 4.441 - (0 . 244^{*} duration of parenteral \\ nutrition) + (2 . 542^{*} KMC more than 3 hours / day) \\ z = 4.441 - (0 . 244^{*} 22) + (2 . 542^{*} 1); \\ z = 1.6 \end{matrix}$

The probability of nosocomial sepsis occurrence in this infant:

P = 1/(1 + e^1.6) = 0.17 (17.0%). During hospital stay this patient didn’t have any incidence of nosocomial sepsis.

Other logistic regression model was built in order to analyze factors that may have an impact on breastfeeding at discharge. Presence of breastfeeding at the moment of discharge from hospital (y = 1) and its absence (formula feeding) (y = 0) were chosen as the binary dependent variables (y). Quantitative and qualitative characteristics that have both promoting and negative influence on the breastfeedingprolongation were chosen as independent variables. Quantitative factors: GA, birth weight, mother’s age, NICU treatment duration, duration of parenteral nutrition. Qualitative characteristic: child’s gender, type of enteral feeding at the hospital (mother’s or donor milk), and long-lasting KMC more than 3 hours per day that has positive influence in promoting breastfeeding. Male gender was coded as “1”, female –“0”, feeding at the hospital with mother’s milk –“1”, with donor milk –“0”; SSC more than 3 hours per day –“1”, KMC less than 3 hours –“0”.

A combination of the NICU treatment duration, type of enteral feeding at the hospital, and mother’s age were identified as factors that had the most important influence on breastfeeding prolongation (Table 4). Other factors that were studied didn’t show the significant indices.

Table 4
Statistical characteristics of the multiple logistic regression factors potentially able to have an impact on breastfeeding at discharge in the study population

Coefficient Standard Error Wald Statistic P value

Constant 13.658 5.233 6.814 0.009

NICU treatment duration 0.119 0.049 5.691 0.017

Type of feeding at hospital –6.358 1.794 12.556 <0.001

Mother’s age –0.370 0.156 5.582 0.018

	Coefficient	Standard Error	Wald Statistic	P value
Constant	13.658	5.233	6.814	0.009
NICU treatment duration	0.119	0.049	5.691	0.017
Type of feeding at hospital	–6.358	1.794	12.556	<0.001
Mother’s age	–0.370	0.156	5.582	0.018

According to the results of the analysis, a multiple regression equation was created: $\begin{matrix} z = 13.658 + (0 . 119^{*} NICU treatment duration) \\ - (6 . 358^{*} type of enteral feeding at hospital) \\ - (0 . 370^{*} mother’s age) . \end{matrix}$

3.4 Examples of calculating the probability of breastfeeding at discharge

Example 3. Female infant, GA 29 + 1, G3P1, birth weight 1290 g (72nd percentile), caesarian section because of maternal bleeding. Mother’s age –34 years. Apgar score 10/10. The NICU treatment duration–8 days, total hospital stay duration –55 days. Baby had RDS, received surfactant therapy, duration of ventilation –5 day, followed by the CPAP-therapy. Infant was fed with mother’s milk during hospital stay, enteral feeding was started on day 1, parenteral nutrition duration –13 days. During hospital stay infant had daily KMCmore than 3 hours/day.

Multiple regression equation was created for this infant: $\begin{matrix} z = 13.658 + (0 . 119^{*} NICU treatment duration) \\ - (6 . 358^{*} type of enteral feeding at hospital) \\ - (0 . 370^{*} {mother}^{'} s age) . \\ z = 13.658 + (0 . 119^{*} 8) - (6 . 358^{*} 1) \\ - (0 . 370^{*} 34); \\ z = - 4.3 \end{matrix}$

The probability of breastfeeding at discharge for this infant:

P = 1/(1 + e^-4.3) = 0.987 (98.7%).

At discharge (55th day of life) infant was exclusively breastfed.

Example 4. Female infant, GA 31 + 0, G6P4, birth weight 1220 g (17th percentile), caesarian section because of maternal bleeding. Mother’s age – 30 years. Apgar score 8/8. Duration of the NICU treatment – 11 days, total hospital stay duration – 38 days. Baby had RDS, received surfactant therapy, duration of ventilation – 1 day, followed by the CPAP-therapy. Infant was fed with donor’s milk during hospital stay, enteral feeding was started on day 0, parenteral nutrition duration – 10 days. During hospital treatment infant had KMC less than 3 hours/day.

Multiple regression equation: $\begin{matrix} z = 13.658 + (0 . 119^{*} NICU treatment duration) \\ - (6 . 358^{*} type of enteral feeding at hospital) \\ - (0 . 370^{*} {mother}^{'} s age) . \\ z = 13.658 + (0 . 119^{*} 11) - (6 . 358^{*} 0) \\ - (0 . 370^{*} 30); \\ z = 3.9 \end{matrix}$

The probability of breastfeeding at discharge for this infant:

P = 1/(1 + e^3.9) = 0.019 (1.9%).

At the moment of discharge (38th day of life) infant had formula feeding.

4 Discussion

Modern perinatal technologies allow to take care of infants with low and extremely low birth weight, preventing the development of most complications associated with premature birth. But nosocomial sepsis in preterm infants is still a severe problem, the incidence ranges from 10 to 30% worldwide [6, 7]. In our study population it was 21%.

Different interventions and measures, such as hand-washing, early enteral feeding, central line care, administration of probiotics and lactofferin are implemented and used in neonatal departments to reduce the incidence of nosocomial infections, but the problem still exists [1, 6]. That is why a great deal of research is devoted to finding effective methods for the nosocomial sepsis prophylaxis.

KMC was found to impact the nosocomial sepsis in preterm infants. Premature newborns who received KMC compared to neonates who have standard care are more likely to have decreased likelihood of nosocomial infection [1, 2]. But most of these studies were performed comparing the results between groups with and without KMC without studying its peculiarities, namely the optimal time of initiating, beneficial duration per day and others.

Feldman et al. pointed out that even one hour of Kangaroo care per day may have positive long-term outcomes [8]. Abouelfettoh et al. stated that 2 hours of KMC is enough to prevent infection [9] however, other researchers investigated that the KMC duration (less and more than 6 hours/day) didn’t show a significant difference in the incidence of sepsis [10]. Our study has shown that prolonged daily KMC, at least more than 3 hours per day has better nosocomial infection prevention results in preterm infants (OR = 10.50; 95% CI 1.23–89.67, p = 0.012).

Nosocomial sepsis is a complex pathological process that is influenced by a variety of factors, both provocative and preventive. Logistic regression analysis identified a combination of factors that significantly influenced nosocomial sepsis occurrence in very preterm infants. It was a combination of parenteral nutrition duration and KMC more than 3 hours per day, as preventing factor. Those babies, who had long duration of catheter-dependent parenteral nutrition, and didn’t have enough KMC, had the highest percentage of nosocomial sepsis occurrence (Example 1). In the same time, those who had regular prolonged KMC more than 3 hours per day and even long-lasting parenteral nutrition, had much more lower risk of nosocomial infection, and actually didn’t have it during hospital stay (Example 2).

Our results are consistent with data of other authors, that parenteral nutrition as a major cause of long-term use of central catheters is a significant factor of nosocomial sepsis in preterm infants. Central line-associated bloodstream infection rank first among the causes of nosocomial sepsis in premature infants [7]. Schulman et al. indicated that the incidence of bacteremia due to central intravascular catheters is 2.1 per 1000 catheter-days [11].

KMC prevents nosocomial infections in preterm infants in different ways. First of all, skin-to-skin contact during KMC promotes the contamination of the baby’s skin by the mother’s non-pathogenic flora, and also reduces the contact time of the child with the medical staff [12]. Cases of pathogenic flora decolonization, methicillin-oxacillin-resistant staphylococci in particular, have also been shown in children who had KMC, indicating the ability of maternal non-pathogenic flora to displace child’spathogenic multi-resistant one [13].

Also skin-to-skin contact during KMC maintains the premature infant’s thermogenesis and prevents the development of hypothermia, which is another significant risk factor of severe bacterial infection [3, 5].

The skin of preterm infants is immature and unable to provide barrier function and protection against pathogens. Skin-to-skin contact during KMC has the ability to improve skin barrier function by reducing trans-epidermal water loss and increasing corneal hydration, thereby reducing the possibility of infection [9].

Kangaroo mother care is accompanied by breastfeeding. Human milk, due to its unique content of antibodies, phagocytes, lactoferrin, and prebiotics improves infant’s immune protection and digestive processes, preventing the occurrence of NEC and sepsis. In addition, complete enteral nutrition starts earlier because of better tolerance to breast milk, reducing the duration of parenteral nutrition and risk of catheter-associated bloodstream infection [2, 14].

Another mechanism is associated with stress and pain reduction during KMC. Premature infants who have long-lasting treatment at hospital are usuallysubjected to stress and have higher levels of circulating cortisol [15]. Increased secretion of corticosteroids reduces the production of lipids and intercellular laminae in the stratum corneum, thus impairing its protective properties and corneal hydrationcapacity. KMC greatly ameliorates stress, reducing the corticosteroid secretion and sympathetic nervous system impact, and thus decreases the risk of stress-induced infections, both skin and systemic [9, 15].

Premature infants management significantly depends on the type of enteral feeding [3, 5]. Breast milk that contains immunological and growth factors, hormones, water-soluble vitamins, antioxidants, and other biologically active substances promotes positive short-term and long-term outcomes for infants [16 –18]. Therefore, it is necessary to support breastfeeding of preterm infants during hospital stay and after discharge.

The Baby-Friendly Hospital initiativestates that KMC with skin-to skin contact is one of the “ten steps to successful breastfeeding” [19]. KMC is proven to be an ideal step towards early and long-lasting exclusive breastfeeding [3, 20]. Various studies reported about higher rates of breastfeeding and its earlier establishment when using both continuous and intermittent KMC compared to the standard care. KMC and non-nutritive sucking increase lactation and promote breastfeeding after discharge [2, 18].

If it is not possible to provide continuous KMC, it is necessary to determine its optimal profitable duration to maintain breastfeeding. Our research confirmed that infants whose KMC duration was more than 3 hours per day were breastfed at discharge significantly more often in comparison with infants who had shorter KMC (OR = 3.7; 95% CI:1.16–11.86; p = 0.025). Some researchers found positive long-term effects of KMC - preterm infants who had more kangaroo care in hospital were breastfed for longer period after discharge [21]. A systematic review reported that even a short period of KMC up to one hour during each visit increased the duration of breastfeeding for more than 6 weeks after discharge [22]. Cochrane meta-analysis pointed out that kangaroo care for twenty minutesprior to feeding promoted the lactation [2].

KMC promotes and supports breastfeeding mainly through oxytocin production. Skin-to-skin contact during KMC stimulates C-afferent nerves on the mother’s breast that cause the release of oxytocin, which plays a significant role in milk ejection reflex [5, 23].

Logistic regression analysis revealed that combination of the NICU treatment duration, type of feeding at the hospital, and mother’s age had the significant impact on breastfeeding rates at discharge in very preterm infants.

NICU treatment of infant has negative influence on mother’s lactation. However, despite the NICU treatment of preterm infant or child’s inability to latch on and suck, mothers should be encouraged to maintain and stimulate lactation by pumping the breast milk. Logistic regression analysis identified feeding with mother’s milk during hospital stay as a significant predictor of breastfeeding at the time of discharge.

Regardless whether the baby sucks the breast or the mother pumps milk, the level of maternal prolactin and oxytocin increases. Prolactin release, associated with infant sucking or milk pumping, maintain the growth of secretory tissue in the mother’s breast for several weeks or months after delivery [18]. Therefore, early, frequent, and effective breastfeeding or, alternatively, milk pumping, is the most important factor in establishing normal lactation [18, 24].

Our study has shown that the rates of breastfeeding at discharge significantly increase with the mother’s age. Our results are similar to that reported by other authors who stated that older mothers and those with lower parity were more likely to exclusively breastfeed at discharge [25], and increased maternal age in combination with multi parity, vaginal delivery and higher education levels were significant positive predictors for early breastfeeding initiation and prolongation [26]. Our research revealed that older mothers provided daily long-lasting KMC for their infants.

KMC is closely related and may have a significant impact on each of the predictors of prolonged breastfeeding. It reduces the NICU treatment duration reduces maternal stress associated with the baby’s stay in the NICU, maintains the lactation during the baby’s stay in hospital, and thus promotes the breastfeeding at discharge. KMC has a positive impact on the mother through the oxytocin production [2 , 18].

5 Conclusions

Thus, the prevention of nosocomial infection in neonatal departments and the optimization of breastfeeding at the time of discharge in very preterm infants are the profitable outcomes of daily prolonged KMC. No significant differences in the BPD, NEC incidences, IVH I–II grades, the duration of parenteral nutrition, and growth parameters were found between groups.

According to logistic regression analysis, a combination of long-lasting KMC (more than 3 hours per day) and short-term parenteral nutrition is a significant factor fornosocomial sepsis prophylaxis in premature babies.

Conflicts of interest

Halyna Pavlyshyn: none.

Iryna Sarapuk: none.

Charlotte Casper: none.

Nataliia Makieieva: none.

Human research statement

Ethics approval was obtained from appropriate local ethics committee and research was conducted in accordance with the World Medical Association’s Helsinki Declaration.

Disclaimer statement

The research was funded by the Ministry of Health of Ukraine by means of the state budget.

References

Campbell-Yeo

, Disher

, Benoit

, Johnston

. Understanding kangaroo care and its benefits to preterm infants. Pediatric Health Med Ther. 2015;6:15–32.

Conde-Agudelo

, Diaz-Rossello

. Kangaroo mother care to reduce morbidity and mortality in low birth weight infants. Cochrane Database Syst Rev. 2014;4:CD002771.

Udani

, Hinduja

, Suman

, Kabra

. Role of kangaroo mother care in preventing neonatal morbidity in the hospital and community: A review article. J Neonatol. 2014;28:29–36.

Cho

, Kim

, Kwon

, Cho

, Kim

, Jun

, et al. The effects of kangaroo care in the neonatal intensive care unit on the physiological functions of preterm infants, maternal-infant attachment, and maternal stress. J Pediatr Nurs. 2016;31:430–8.

Ludington-Hoe

. Skin-to-skin contact: A comforting place with comfort food. MCN Am J Matern Child Nurs. 2015;40:359–66.

Ramasethu

. Prevention and treatment of neonatal nosocomial infections. Matern Health Neonatol Perinatol. 2017;3:5.

Nanou

, Paulopoulou

, Losis

, Tsoumakas

, Saroglou

. Risk factors for nosocomial infections in neonatal intensive care units (NICU). Health Sci J. 2015;9:9.

Feldman

, Rosenthal

, Eidelman

. Maternal-preterm skin-to-skin contact enhances child physiologic organization and cognitive control across the first 10 years of life. Biol Psychiatry. 2014;75:56–64.

Abouelfettoh

, Ludington-Hoe

, Burant

, Visscher

. Effect of skin-to-skin contact on preterm infant skin barrier function and hospital-acquired infection. J Clin Med Res. 2011;3:36–46.

10.

Udani

, Kabra

, Nanavati

, Aloke

. Kangaroo Mother Care (KMC): A cohort follow up study on impact of duration of KMC on mortality, morbidity, hospital stay and breastfeeding. J Neonatol. 2013;27:5–10.

11.

Schulman

, Stricof

, Stevens

, Horgan

, Gase

, Holzman

, et al. Statewide NICU central-line-associated bloodstream infection rates decline after bundles and checklists. Pediatrics. 2011;127:436–44.

12.

Lawn

, Mwansa-Kambafwile

, Horta

, Barros

, Cousens

. ‘Kangaroo mother care’ to prevent neonatal deaths due to preterm birth complications. Int J Epidemiol. 2010;39:144–54.

13.

Filho

, Cavalcante de Sousa

, Freitas

, Lamy

, Simoes

, da Silva

, et al. Effect of maternal skin-to-skin contact on decolonization of methicillin-oxacillin-resistant staphylococcus in neonatal intensive care units: A randomized controlled trial. BMC Pregnancy Childbirth. 2015;15:63–9.

14.

Patel

, Johnson

, Engstrom

, Fogg

, Jegier

, Bigger

, et al. Impact of early human milk on sepsis and health care costs in very low birth weight infants. J Perinatol. 2013;33:514–9.

15.

Morelius

, Theodorsson

, Nelson

. Salivary cortisoland mood and pain profiles during skin-to-skin care for an unselected group of mothers and infants in neonatal intensive care. Pediatrics. 2005;116:1105–113.

16.

Dicky

, Ehlinger

, Montjaux

, Gremmo-Feger

, Sizun

, Roze

, et al. Policy of feeding very preterm infants with their mother’s own fresh expressed milk was associated with a reduced risk of bronchopulmonary dysplasia. Acta Paediatr. 2017;106:755–62.

17.

American Academy of Pediatrics Section on Breast feeding . Policy statement: Breast feeding and the use of human milk. Pediatrics. 2012;129:e827–e41.

18.

Wight

. Breast feeding the NICU infant: What to expect. Clin Obstet Gynecol. 2015;58:840–54.

19.

Baby-Friendly Hospital Initiative: Revised, updated and expanded for integrated care. Geneva: World Health Organization; 2009.

20.

Baley

. Skin-to-skin care for term and preterm infants in the neonatal ICU. Pediatrics. 2015;136:596–9.

21.

Flacking

, Ewald

, Wallin

. Positive effect of kangaroo mother care on long-term breastfeeding in very preterm infants. J Obstet Gynecol Neonatal Nurs. 2011;40:190–97.

22.

Renfrew

, Craig

, Dyson

, McCormick

, Rice

, King

, et al. Breast feeding promotion for infants in neonatal units: a systematic review and economic analysis. Health Technol Assess. 2009;13:1–146.

23.

Uvnäs-Moberg

, Prime

. Oxytocin effects in mothers and infants during breastfeeding. Infant. 2013;9:201–6.

24.

Neville

, Morton

, Umemura

. Lactogenesis. Pediatr Clin North Am. 2001;48:35–52.

25.

Hundalani

, Irigoyen

, Braitman

, Matam

, Mandakovic-Falconi

. Breastfeeding among inner-city women: from intention before delivery to breast feeding at hospital discharge. Breastfeed Med. 2013;8:68–72.

26.

Amin

, Hablas

, Al Qader

. Determinants of initiation and exclusivity of breast feeding in Al Hassa, Saudi Arabia. Breastfeed Med. 2011;6:59–68.

	Group 1	Group 2
	KMC more than 3 hours/day	KMC less than 3 hours/day	p value
	(n = 22)	(n = 30)
Maternal clinical characteristics
Maternal age (years), Me [Lq; Uq]	33.0 [30.0; 36.0]	29.0 [26.0; 33.0]^*	0.009
Gestational hypertension, preeclampsia/eclampsia, n (%)	7 (31.8)	12 (40.0)	0.544
Maternal infection, n (%)	8 (36.4)	9 (30.0)	0.632
Antenatal steroid, n (%)	20 (90.9)	26 (86.7)	0.639
Maternal hemorrhage, n (%)	3 (13.6)	6 (20.0)	0.547
Fetal hypoxia, n (%)	8 (36.4)	10 (33.3)	0.822
Characteristics of newborns
Gestation age, Me [Lq; Uq]	29.92 [29.14; 31.57]	30.43 [28.86; 31.28]	0.833
Birth weight (grams), Me [Lq; Uq]	1222.5 [1050.0; 1530.0]	1270.0 [1020.0; 1480.0]	0.920
Birth weight (percentile), Me [Lq;Uq]	30.5 [8.0; 53.0]	25.5 [7.0; 54.0]	0.949
Small for gestational age, n (%)	6 (27.3)	7 (23.3)	0.742
Apgar score at 1 min, Me [Lq; Uq]	9.0 [8.0; 10.0]	7.5 [5.0; 9.0]	0.057
Apgar 5 score at 5 min, Me [Lq; Uq]	10.0 [9.0;10.0]	10.0 [8.0; 10.0]	0.220
Primary resuscitation with ventilation, n (%)	11 (50.0)	9 (30.0)	0.143
Heart massage and epinephrine administration, n (%)	1 (4.5)	0 (0)	0.241
Respiratory distress syndrome (RDS), n (%)	10 (45.5)	16 (53.3)	0.578
Early-onset sepsis, n (%)	3 (13.6)	3 (10.0)	0.688
NICU treatment duration (days)	9.5 [7.0; 12.0]	7.0 [5.0; 15.0]	0.462