The Possible Beneficial Effect of Breastfeeding on the Clinical Course of Urolithiasis Detected During Infancy

Abstract

Aim:

To evaluate the possible effect of breastfeeding duration on the clinical course and treatment of stones detected during infancy.

Materials and Methods:

Forty-eight infants with renal stones diagnosed between 0 and 23 months were included in this retrospective cohort study. The children were called for regular follow-up visits for the evaluation of physical examination findings, renal size and parenchymal thickness measured by ultrasonography, localization, size and number of stones, spot urine analysis data, metabolic evaluation findings, and duration of breastfeeding and formula use. The relationship between the disease course (progression or stability) and the duration of breastfeeding were assessed from different aspects by considering the original characteristics of the stones at the time of first diagnosis.

Results:

The mean follow-up period was 46.21 ± 23.22 months, and the most important metabolic risk factor was hypercalciuria with a rate of 29.2%. The mean duration of breast milk intake was longer in children with no disease progression and in those with a reduced size and/or number of stones. The children receiving only breast milk for the first 6 months of life required treatment less and had less growth retardation.

Conclusions:

Our findings suggest that breastfeeding has critical effects on infantile urolithiasis. Breastfeeding should be encouraged in children with stones particularly detected during infancy. Further studies with a larger case series are needed to reveal the positive effects of breast milk on the clinical course of urolithiasis in children.

Introduction

Pediatric urolithiasis has been reported to be a rare condition with an overall incidence of 1–2%.¹ However, because various factors can cause significant complications in the kidneys, such as renal failure, fluid–electrolyte imbalance, and growth retardation, impairing the general condition of the patient, it is important to closely monitor all infants with stones, and to carry out a thorough assessment.²

Taking the possible risk of harmful effects in the kidneys, and the metabolic status related to persistent or recurrent stone formation, a close follow-up is very important for the clinical progress of stones in infants to achieve an early diagnosis and plan timely appropriate treatment. A review of the literature clearly shows the availability of very limited reliable data regarding the incidence, metabolic abnormalities, recurrence rates, and important factors affecting the natural course of the disease in pediatric cases compared with adults.^3,4 Thus, considering the high likelihood of the above-mentioned adverse effects of the stones on the growing body and the kidneys of infants, the examination of patients and disease-related factors requires careful attention and thorough follow-up for all children after the first stone attack.

It is known that the type of diet, fluid intake, and medications, for example, hypervitaminosis D because of excessive vitamin D intake or calcium supplements are important disorders that can lead to urinary stones.⁵ The type of diet is an important factor especially during postinfancy, and infantile urolithiasis is also related to genetic inheritance and metabolic disorders. In the first 6 months of life, duration and amount of breastfeeding are important to reduce the risk of urinary tract infection (UTI), growth retardation, and many other conditions impairing kidney functions.^6–8 To the best of our knowledge, this is the first study that analyzed the relationship between breastfeeding duration and the clinical course of urinary stones in infants. There are also very limited data in the literature about the possible effects of the use of formula and/or breast milk over a long term as in our study with a follow-up period of ∼4 years.

In this study, we aimed to evaluate the metabolic factors and the clinical course of the stones detected during infancy to assess the possible outcome of the disease and the relationship between breastfeeding duration and progression of the disease.

Materials and Methods

A total of 48 infants (<24 months old) (31 boys and 17 girls; M/F: 1.8) diagnosed with primary urinary stones by renal ultrasonography were included in the study, and their medical files were evaluated. The test findings of the metabolic evaluation were noted based on serum- and urine-related risk factors basically depending on spot urine analysis. Infants demonstrating anatomic abnormalities, neurogenic disorders, or urinary diversions, those who were born premature, and those receiving any diuretic agents, especially furosemide, were excluded from the study.

To ensure the presence of stone(s) in the kidney, if there was any suspicion related to the diagnosis of urinary stones (e.g., possibility of nephrocalcinosis), then these cases were also excluded. In addition, infants with heart failure, renal failure, anomalies in the genitourinary system, and indications for surgical intervention at the time of diagnosis, and those who were taking any drug that could affect mineral metabolism, including vitamin D intake of >400 IU were not included in the study. Depending on the size of the stones diagnosed and factors related to urinary tract obstruction, infection, and hematuria requiring an urgent intervention, some cases were also excluded from the follow-up program.

The clinical course of the stone disease in all infants was well assessed and noted. In addition to the age at first presentation, the stone size, presence of positive family history, and results of metabolic evaluation, including the feeding status of the children during the follow-up (total duration of breast feeding and use of formula) were carefully recorded. All children were also well evaluated regarding the presence and frequency of UTIs, interventions performed for stone removal, and more importantly physical and renal growth rates and renal parenchymal thickness measurements during this relatively long follow-up period by the same physicians and radiologist using the same method (renal ultrasound and/or plain radiography).

After obtaining the patient history, a careful physical examination and biochemical analysis, including urinalysis and culture sensitivity tests, were performed in all infants. In the majority of the cases, a plain film and/or urinary sonography were the initial radiological methods to image the stone. Growth retardation was defined as a height Z score of 2 standard deviation (SD) below the mean value for age and sex, and/or 2 SD below mid-parenteral height SD. All cases of growth retardation were identified based on a weight percentile <3.⁹

The measurement of renal size (length and width) and parenchymal thickness values was undertaken using renal sonography and comparatively evaluated with respect to the nomograms of renal parenchymal thickness and renal length in healthy children. Reduced renal size and/or renal parenchymal thickness were defined with the values obtained <2 SD.¹⁰ In addition to spontaneous passage and persistence of stones, regrowth of stones and new stone formation were well assessed in each case.

Before beginning the study, the study protocol was approved by the regional ethical committee (80576354-050-99/16; December 26, 2018), and written informed consent was obtained from the parents.

The Statistical Package for the Social Sciences v. 20 (SPSS, IL) was used for statistical analysis. In addition to the definitive statistical methods (mean, SD, median, frequency, rate, minimum and maximum) used during the evaluation of the data obtained, the Mann–Whitney U test was conducted for the evaluation of abnormally distributed parameters between two groups to compare the quantitative data. Furthermore, Pearson's chi-square and Fisher's exact tests were performed for the evaluation and comparison of qualitative parameters between the groups. Statistical significance was defined as p < 0.01 and p < 0.05.

Results

The follow-up duration ranged from 9 to 110 months (mean = 46.21 ± 23.22 months). A total of 48 infants (31 boys and 17 girls, M/F: 1.8, age 18–132 months) with a true diagnosis of kidney stone(s) based on appropriate imaging modalities were included in the study. At the time of diagnosis, all patients were in good health conditions, and their vital signs were stable. The patients included in the study all met the above-mentioned criteria, had microlithiasis and/or macrolithiasis in their urinary systems, did not have renal failure or an indication for surgical treatment, and did not require treatment for infections.

The patient demographics and clinical features are summarized in Table 1. Of the infants, 52.1% had more than one stone and 47.9% had relatively large calculi (macrolithiasis). Although 18.8% of the cases needed surgical treatment for these stones, spontaneous remission was noted during the follow-up in 62.5% of the cases. The most common metabolic risk factor was hypercalciuria detected in 29.2% of the cases evaluated. Stone analysis was not possible in all cases because of the difficulties in obtaining and performing 24-hour urine analysis in this highly specific population, but the spot urine examination revealed calcium oxalate in stone composition in the vast majority of children.

Table 1.

Patients' Demographic and Clinical Features

Age (months; min–max)	58.15 ± 29.017	18–132
Age at first diagnosis (months; min–max)	10.52 ± 7.84	15 days–23 months
Female (n, %)	17	35.4
Male (n, %)	31	64.6
Duration of follow-up (months; min–max)	46.21 ± 23.22	9–110
Duration of breastfeeding (months; min–max)	18.38 ± 8.40	0–30
Duration of feeding with formula (months; min–max)	3.46 ± 7.61	0–36
Family history of nephrolithiasis (n, %)	21	43.8
Growth retardation (n, %)	13	27.1
UTI during follow-up (n, %)
None	40	83.6
1	3	6.3
>3	5	10.1

Values are given as mean ± SD unless otherwise specified.

UTI, urinary tract infection.

Table 2 demonstrates the evaluation data related to the metabolic risk factors and features of stones. The children were divided into two groups depending on the progression of urinary stone disease during the follow-up. The first group consisted of cases in remission with reduced size and number of stones, and the second group comprised those with persistent disease presenting with increased size and number of stones. When we compared the breastfeeding status, growth retardation, presence of UTI attacks, and family history between the two groups, we determined that the duration of breastfeeding was significantly longer and the duration of formula use was significantly lower in the first group, whereas a family history of nephrolithiasis was significantly higher in the second group during the follow-up.

Table 2.

Evaluation of Stones and Metabolic Risk Factors

	n	%
No. of stones
1	23	47.9
>1	25	52.1
Treatment need
Only follow-up	35	72.9
Medical treatment	4	8.3
Surgical treatment	9	18.8
Final stone status
Remission	33	68.9
Decreased number and size of stones	5	10.4
Persistent stones or increased number and size of stones	10	20.9
Metabolic risk factors (45/48)
Hypercalciuria	14	29.2
Hypocitraturia	3	6.3
Hyperoxaluria	3	6.3
Hyperuricosuria	0
Hypomagnesuria	0

Table 3 summarizes the clinical course of the stones in all groups in a comparative manner based on the duration of breastfeeding period and risk factors. According to the feeding status, the patients were further divided into three groups: children who were only breastfed (breast milk group), those who were only fed with formula without any breast milk intake (formula group), and those receiving both formula and breast milk (breast milk + formula group) during the first 6 months of life.

Table 3.

Evaluation of Some Risk Factors According to Stone Disease Progression

	Remission or decreased size and number of stones	Persistent stones or increased size and number of stones	p
Duration of breastfeeding (months) (mean ± SD)	20.47 ± 7.18	10.40 ± 8.22	0.000
Duration of formula use (months) (mean ± SD)	6.39 ± 2.05	8.80 ± 9.71	0.011
UTI (%, n)
None	71.1	80
1	18.4	10	0.808
>3	10.5	10
Family history of nephrolithiasis (%, n)			0.047
Yes	34.2	70
No	65.8	30
Treatment requirement (%, n)
None	78.9	50
Medical	5.3	20	0.148
Surgical	15.8	30
Growth retardation (%, n)
Yes	23.7	40	0.257
No	76.3	60

Bold value indicates p < 0.05 is significant.

UTI, urinary tract infection.

There was no significant difference between these three groups with respect to the renal size, renal parenchymal thickness measurement, and frequency of UTI attacks during the follow-up. However, in the breast milk group, the rate of disease remission and/or reduced size and number of stones was significantly higher (p = 0.028, p < 0.05) and growth retardation was less likely (limited) (p = 0.018), whereas in the formula group, the requirement of treatment was significantly higher (p = 0.016) (Table 4). There was no significant relationship between the mother's social status and the type of diet that may affect the milk content (Table 5).

Table 4.

Evaluation of Some Risk Factors and Progression of the Stones According to Type of Feeding in the First 6 Months of Life

	Breast milk^a	Formula^b	Breast milk + formula^c	p
Progression of stone disease
Remission and/or decrease size and number of stones (%)	88.6	60	50	0.028
Persistent stones and/or increased size and number of stones (%)	11.4	40	50
Treatment requirement (%)
None	77.1	20	75	0.016
Medical and/or surgical	22.9	80	25
Renal size (%)
Normal (45; 93.6%)	91.4	100	100	0.552
Reduced (3; 6.3%)	8.6	0	0
Renal parenchymal thickness (%)
Normal (30; 62.5%)	54.3	80	87.5	0.150
Reduced (18; 37.5%)	45.7	20	12.5
Growth retardation
Yes	20	80	25	0.018
No	80	20	75
UTI (%)
None	74.3	40	75
1	17.1	20	12.5	0.181
>3	8.6	40	12.5

Group 1.

Group 2.

Group 3.

Bold value indicates p < 0.05 is significant.

UTI, urinary tract infection.

Table 5.

Sociodemographic Features That May Affect the Breast Milk Production of Mothers

	Group 1	Group 2	Group 3	p
Area of residence (%)
Rural	51.4	80	50	0.571
Urban	48.6	20	50
Income (%)
Middle–high	71.4	60	75	0.879
Little	28.6	40	25
Mother's diet (%)
Vegan	0	0	0
Based on milk and milk production	20	0	2	0.907
Based on protein	25.7	40	50
Healthy, balanced diet	54.3	60	50

Discussion

Despite its relatively rare occurrence (1–2%),¹ infantile urolithiasis has been reported to remain an important problem in Iran, Turkey, Armenia, Iraq, and some other European countries.¹¹ Stone composition has been found to be mainly related to socioeconomic conditions, geography, and dietary habits,¹² and the data published to date have demonstrated that >75% of all children have calcium-containing (oxalate or phosphate) calculi.¹³ The disease can also lead to some serious risks and harmful effects on renal function because of the recurrent nature of the problem in this specific population.¹⁴ Therefore, it is crucial to perform appropriate and timely management of the disease by completely removing stones and limiting new stone formation, eradicating UTIs, and treating both anatomic abnormalities and underlying metabolic disorders.¹⁵

In the management of infantile urolithiasis, the options include follow-up alone, medical treatment, and surgery according to the patient's risk factors and clinical and laboratory findings. Although the stones of some patients regress without treatment, the disease may progress despite medical and/or surgical treatment in others.^3,15 In our study, the mean age of the children was 58.15 ± 29.017 months, and all were diagnosed with renal stones during infancy. We closely followed up the cases over a period ranging from 9 to 110 months. There was no recurrence in any of the patients, but persistent disease and/or increased number and size of stones with or without treatment were seen in 20.9% of the cases. Because of the high risk of recurrent attacks and/or persistent stones, which is associated with considerable morbidity, a long-term postoperative follow-up is mandatory.¹⁶

In the literature, one study reported a 4.2% recurrence rate in 91 children during a mean follow-up of 38.2 months.¹⁷ Another study showed that recurrence occurred in only 2% of cases in a large group of children during a 13-year follow-up period.¹⁸ Without any treatment, the rate of infantile urolithiasis recurrence and/or persistence can reach 50%.^16–18 However, unlike adult cases, there are only limited data published in the literature focusing on the clinical course of stone disease in children particularly in the long-term.¹⁹

In this study, growth retardation and UTI (single or recurrent) that can have a possible additional adverse effect on the clinical progress of stone disease did not seem to be significantly related to the long-term outcome of the children evaluated. Only 16.4% of the children in our sample had a UTI attack during the follow-up period; however, we believe that a larger case series is needed to better outline the true incidence of UTI in this patient group.

Genetic inheritance is another risk factor for urolithiasis, especially in the infantile period. A positive family history of nephrolithiasis is an important factor requiring careful evaluation. In this regard, although a study conducted in Mexico²⁰ revealed that a positive family history of nephrolithiasis was present in 50% of cases, another study from Albania²¹ showed that this rate was 45.8%. In this study, we found the presence of positive family history in 43.8% of the children, which is consistent with the literature.

Metabolic disorders are considered to be the most important factors among the causes of infantile urolithiasis. Accumulated data in the published literature show that hypercalciuria, hyperoxaluria, and hypocitraturia are most commonly detected urinary risk factors. In this study, the most frequent metabolic disorder was hypercalciuria, being detected in 29.2% of the infants. Milošević et al. showed that the most common metabolic risk factor was hypercalciuria at a rate of 47.37%,²² whereas Bozkurt et al. reported it to be hypercalciuria at a rate of 25.4%.²³ Hypercalciuria may be associated with the intake of excessive vitamin D or calcium.²⁴ We excluded children with an excessive vitamin D or calcium intake; therefore, we consider that the frequency of hypercalciuria in the sample may have been related to the children's predisposition to the hereditary form of the condition.

In addition to the intake of excessive vitamin D, the type of diet is an important factor associated with the etiology of urinary stones in infants. For example, high protein intake can also lead to urolithiasis by increasing the urinary uric acid concentration and decreasing the urinary citrate concentration.²⁴ Some studies showed that a fat-rich diet and obesity could be associated with renal stone formation, whereas others presented malnutrition as an important factor in a subset of children forming stones.^9,25

In this study, 27.1% of patients had growth retardation related to the low intake of breast milk. Similarly, Sarıca et al.,¹⁰ who evaluated children diagnosed with stones during the early childhood period, demonstrated a growth retardation rate of 22% in cases where the breastfeeding duration was lower, but this was not significant. Although feeding type and diet are crucial in all age groups, these factors are particularly important during the first 6 months of life, where breast milk constitutes the most essential part of diet.

The World Health Organization recommended providing infant with exclusive breast milk for the first 6 months of life, and if possible extending this period up to 24 months or even longer.²⁶ Formula is only an acceptable alternative when breast milk is inadequate or if the baby cannot consume breast milk for various reasons. It has been shown that breast milk is not only an important dietary component, but it also acts as the first vaccine and probiotic in preventing infections and promoting mental development.²⁷ In underdeveloped countries, it helps prevent malnutrition and associated risks, such as overweight and diabetes.^8,28

A possible relationship between breastfeeding and urolithiasis has been reported with limited data in recent years. Based on the findings of our study, we can state that the longer duration of breastfeeding contributes to the clinical regression of the urinary stone disease. Exclusive breast milk intake in the first 6 months of life seems to be very important for lowering the need for additional treatment. Sarıca et al. also showed that the duration of breastfeeding was lower in cases with retarded growth presenting with a reduced renal size and parenchymal thickness, although their results were not significant.²⁵ In our study, we did not determine a significant relationship between breastfeeding duration and renal size and parenchymal thickness, but the rate of growth retardation was significantly lower in the group fed exclusively with breast milk during the first 6 months of life.

The amount of calcium and phosphorus in breast milk does not change much during lactation, and although it is lower in breast milk than in formula, the bioavailability of the former is better because of its high absorption. Another important point is that the ratio of calcium is ∼35 mg/dL in breast milk and 50–55 mg/dL in formula. Because the calcium:phosphorus ratio in breast milk is ∼2:1, it has higher absorption and is more suitable for bone mineralization.²⁹ Therefore, calcium is not required in term infants who are breastfed unless there is an underlying pathological condition, hypercalcemia and hypercalciuria that play a role in stone etiology are also not expected in these children. On the contrary, it is known that high phosphorus (35–42 mg/dL) and high protein ratios in formula increase renal solute load.

However, a study conducted in China provided no evidence that the routine use of formula, except for melamine-contaminated formula, caused urolithiasis.³⁰ In the light of this information, rather than stating that formula predisposes infants to urolithiasis, it would be a more accurate to conclude that breast milk intake positively affects the prognosis of urolithiasis.

Considering the highly limited data reported in the literature particularly in infants with stone disease, we believe that the findings obtained from this study contribute to our knowledge of the natural clinical course of stone disease in these cases. Our results clearly showed that the duration of sole breastfeeding affected the course of stone disease in infants starting from birth and a longer duration resulted in significantly decreased stone size and/or numbers. Thus, the duration of breastfeeding seems to be highly related to events that are associated with the development of infantile urolithiasis. The possible preventive effect of breast milk on UTIs can be responsible for the clinical course of urinary stones. Ajetunmobi et al. showed that the risk of long hospitalization and UTI formation was higher among formula-fed infants compared with breastfed infants.³¹ Some breast milk components, such as lactoferrin, immunoglobulin, and antiadhesive oligosaccharides can provide this protective effect. In this study, we found that the patients referred to the clinic with more than three UTI attacks were more frequently included in the formula group compared with the cases in the breast milk and breast milk + formula groups. However, there was no significant difference between these three groups.

Limitations

This study has certain limitations. First, our study group had a relatively small number of children. Second, the mothers' socioeconomic status was not evaluated, although it can possibly affect nutrition and breast milk content. However, considering the highly limited data reported in the literature to date, especially in this specific and rare patient population, we believe that our findings make a considerable contribution to the understanding of this critical subject.

Conclusion

The literature data published to date suggest that all children with urolithiasis should be carefully and regularly followed up by pediatricians and pediatric urologists through detailed physical and radiological examinations. Breastfeeding duration and the exclusive intake of breast milk in the first 6 months of age affect the progression of urinary stone. Therefore, we suggest that breastfeeding should be encouraged in infants with urolithiasis. However, more detailed studies with a larger number of patients are needed to reveal the effects of breast milk on urolithiasis in children.

Disclosure Statement

For all authors, no competing financial interests exist.

Funding Information

No funding was received for this article.

References

Jackson

, Avendt-Reeber

. Urolithiasis in children—Treatment and prevention. Curr Treat Options Pediatr, 2016; 2:10–22.

Spivacow

, Negri

, Del Valle

, et al. Metabolic risk factorsin children with kidney stone disease. Pediatr Nephrol, 2008; 23:1129–1133.

Hernandez

, Ellison

, Lendvay

. Current trends, evaluation, and management of pediatric nephrolithiasis. JAMA Pediatr, 2015; 169:964–970.

Koyuncu

, Yencilek

, Erturhan

, et al. Clinical course of pediatric urolithiasis: Follow-up data in a long-term basis. Int Urol Nephrol, 2011; 43:7–13.

Bastug

, Gunduz

, Tulpar

, et al. Urolithiasis in infants: Evaluation of risk factors. World J Urol, 2013; 31:1117–1122.

Miliku

, Voortman

, Bakker

, et al. Infant breastfeeding and kidney function in schoolaged children. Am J Kidney Dis, 2015; 66:421–428.

Hanson

. Protective effects of breastfeeding against urinary tract infection. Acta Pædiatr, 2004; 93:154–156.

Scherbaum

, Srour

. The role of breastfeeding in the prevention of childhood malnutrition. World Rev Nutr Diet, 2016; 115:82–97.

Kim

, Luan

, Canning

, et al. Association between body mass index and urolithiasis in children. J Urol, 2011; 186 [Suppl]:1734–1739.

10.

Sarica

, Narter

, Sabuncu

. Factors affecting the course of body and kidney growth in infants with urolithiasis: A critical long-term evaluation. Arch Ital Urol Androl, 2016; 88:249–254.

11.

Alpay

, Gokce

, Özen

, et al. Urinary stone disease in the first year of life: Is it dangerous?. Pediatr Surg Int, 2013; 29:311–316.

12.

Wang

, Wiener

, Lipkin

, et al. Estimating the nationwide, hospital based economic impact of pediatric urolithiasis. J Urol, 2015; 193(5 suppl):1855–1859.

13.

Bak

, Ural

, Agin

, et al. The metabolic etiology of urolithiasisin Turkish children. Int Urol Nephrol, 2009; l41:453–460.

14.

Alpay

, Ozen

, Gokce

, et al. Clinical and metabolic features of urolithiasis and microlithiasis in children. Pediatr Nephrol, 2009; 24:2203–2209.

15.

Guven

, Koyun

, Baysal

, et al. Urolithiasis in the first year of life. Pediatr Nephrol, 2010; 25:129–134.

16.

Schwarz

, Dwyer

. Pediatric kidney stones: Long-term outcomes. Urology, 2006; 67:812–816.

17.

Erbagcı

, Erbagcı

, Yılmaz

, et al. Pediatric urolithiasis. Scand J Urol Nephrol, 2003; 37:129–133.

18.

Rizvi

, Naqvi

, Hussain

, et al. Pediatric urolithiasis: Developing nation perspectives. J Urol, 2002; 168:1522–1525.

19.

Marzuillo

, Guarino

, Apicella

, et al. Why we need a higher suspicion index of urolithiasis in children. J Pediatr Urol, 2017; 13:164–171.

20.

Velásquez-Forero

, Esparza

, Salas

, et al. Risk factors evaluation for urolithiasis among children (in Spanish). Bol Med Hosp Infant Mex, 2016; 73:228–236.

21.

Shıtza

, Shukrti

, Pistulli

, et al. Urolithiasis in children—Albanian experience. Int J Sci Res, 2016; 1:1214–1219.

22.

Milošević

, Batinić

, Turudić

, et al. Demographic characteristics and metabolic risk factors in Croatian children with urolithiasis. Eur J Pediatr, 2014; 173: 353–359.

23.

Bozkurt

, Ece

, Yolbaş

. et al. Pediatric urolithiasis: An evaluation of 342 cases (in Turkish). Düzce Tıp Fakültesi Dergis, 2009; 11:21–25.

24.

Baştuğ

, Düşünsel

. Pediatric urolithiasis: Causative factors, diagnosis and medical management. Nat Rev Urol, 2012; 9:138–146.

25.

Selimoğlu

, Menekşe

, Tabel

. Is urolithiasis in children associated with obesity or malnutrition?. J Ren Nutr, 2013; 23:119–122.

26.

World Health Organization. Breastfeeeding-Recommendation. https://www.paho.org/hq/index.php?option=com_content&view=article&id=9327:breastfeeding-recommendations&Itemid=42404&lang=en (accessed July 15, 2019).

27.

Victora

, Bahl R

, arros

, et al.; The Lancet Breastfeeding Series Group. Breastfeeding in the 21st century: Epidemiology, mechanisms, and lifelong effect. Lancet, 2016; 387:475–490.

28.

Horta

, de Mola

, Victora

. Long-term consequences of breastfeeding on cholesterol, obesity, systolic blood pressure, and type-2 diabetes: Systematic review and meta-analysis. Acta Paediatr Suppl, 2015; 104:30–37.

29.

Venkataraman

, Luhar

, Neylan

. Bone mineral metabolism in full-term infants fed human milk, cow milk-based, and soy-based formulas. Am J Dis Child, 1992; 146:1302.

30.

Chang

, Wu

, Yue

, et al. Melamine poisoning pediatric urolithiasis treatment in Gansu, China 5-year follow-up analysis. Urology, 2017; 109:153–158.

31.

Ajetunmobi

, Whyte

, Chalmers

, et al. Breastfeeding is associated with reduced childhood hospitalization: Evidence from Scottish Birth Cohort (1997–2009). J Pediatr, 2015; 166:620–625.