Preterm infants are at risk of decreased bone strength and mineralization due to limited accretion of bone mass in utero [1,2]. Significant bone demineralization is seen at 40 weeks in more than 50% of infants with birthweight <1500 g and almost all infants with birthweight of less than 1000g [3,4]. Osteopenia is exacerbated by prolonged parenteral nutrition, postnatal morbidity like bronchopulmonary dysplasia and necrotizing enterocolitis, and medications like diuretics, steroids and caffeine [5]. Limited in utero accretion is further compounded by inactivity because of inherent weakness, poor tone and confinement (nesting) leading to increase in bone resorption, and decrease in bone mineral density [6]. Previous studies that employed physical exercise while in hospital for short period (4 to 8 weeks) had demonstrated increased bone mineralization and growth in preterm infants [7].

This open-label randomized controlled trial was carried out in a level-3 neonatal unit in a tertiary-care teaching hospital in northern India between 16 May, 2013 and 21 November, 2013. Informed written consent was obtained from one of the parents before enrolment of infants and the study was approved by the Institutional Ethics Committee. All preterm infants born between 270/7 and 346/7 weeks of gestation were eligible for study at one week of postnatal age. Infants with major malformations, intraventricular hemorrhage grade 3 or 4, necrotizing enterocolitis stage IIB or more, any abdominal surgery, shock needing inotropes after first week, those needing high frequency ventilation after first week, or suspected bony dysplasia were excluded. Gestational age (GA) was ascertained from the first day of last menstrual period or by the ultrasound at first trimester or by the Expanded New Ballard Score [8] (ENBS) performed within 24 hours of birth in that order of preference. Eligible neonates were stratified by gestation age into two strata: (27^0/7 to 32^6/7 weeks) and (33^0/7 to 34^6/7 weeks). Infants in both the strata were randomized to Exercise group or Control group. We used stratified block randomization with a variable block size of 2 to 8, and used computer-generated random numbers for random sequence generation. Serially numbered, opaque and sealed envelopes were used to ensure allocation concealment. These envelopes were opened in front of parents after obtaining informed consent at attainment of one week of postnatal age. Blinding was not feasible due to the nature of the intervention. However, the assessors measuring all outcomes were blinded to the group assignment.

Infants were fed expressed breast milk (EBM) by gavage or spoon. EBM was fortified with human milk fortifier (Lactodex-HMF, Raptakos, Brett & Co. Ltd, India) once the infant was on at least 100 mL/kg feeds as per the existing unit protocol [10]. If EBM was not available, preterm formula was used. At discharge, mothers were advised to continue HMF until the infant attained a body weight of 2000 g. All the infants were also supplemented with 800 IU of vitamin D3 daily initially, and later changed to 400 IU after a change in unit protocol [11]. Routine Iron (2 mg/kg/day) and Vitamin D3 supplements were given to all infants at discharge. The calculation for vitamin D, Calcium and phosphorus were done as an average of daily intake by calculating separately the intake through milk, HMF and additional vitamin D3 drops.

The primary outcome was the Speed of sound (SOS) of left tibial bone by quantitative ultrasound (QUS) Sunlight Omnisense 7000P/8000P (BeamMed Ltd, Israel) as a surrogate marker of bone strength. The principal investigator measured the baseline SOS and a neonatologist blinded to group assignment measured the primary outcome. Both the principal investigator and the primary outcome assessor at 40 weeks were trained by another expert clinician in the department and the service engineer of the quantitative ultrasound machine. Before initiating the study, the inter-observer reliability, as assessed by the intraclass correlation (ICC) was 0.89 (95% CI 0.82 to 0.96). Before each measurement, a system quality verification was done by using the phantom provided by the manufacturer. If the reading remained in the narrow green zone, the probe was considered to be functioning properly. The distance from the knee to the heel of left tibia was measured using a calliper and one half of the distance was noted and marked by a pencil. After application of ultrasound gel on the probe, it was moved over the medial aspect of the mid shaft tibia at the marked point for maximal reading to obtain measurements of SOS and SOS Z score. The mean of best three measurements was selected for data analysis and this calculation was entirely equipment dependent. Change in SOS (DSOS) was defined as the difference between baseline and 40 weeks SOS measurements.

A total of 53 infants were born between 27+0 and 34+6 weeks of gestation during the study period. We enrolled 50 neonates (n=26 in EG and n=24 in CG) in the study, at attainment of completion of one week of age, out of which 25 and 22 infants were available for analysis at 40 weeks in EG and CG groups, respectively (Fig 1). The baseline and other clinical characteristics of the two study groups were comparable (Table I). The exercise sessions were undertaken on a mean (SD) 48.7 (16.3) days. Overall compliance of daily exercise was 94.6%. There were no adverse events in form of destabilization of vitals or fracture of bones throughout our study.

Fig.1 Study flow chart.

TABLE I	Baseline Characteristics of the Enrolled Infants

TABLE II Bone Speed of Sound at 40 week Post-menstrual Age in Enrolled Infants (n=47) (at 40 Weeks)

Fig. 2 Decline in bone speed of sound (m/sec) from baseline (at enrolment) to term age.

There was no difference in tibial SOS at 40 weeks between the groups [2858 (142) vs. 2791 (122); mean difference 67.6 m/sec; (95% CI - 11 to 146; P=0.38)] (Table II). Even after adjustment for birthweight, gender, duration of parenteral nutrition, and baseline bone SOS, the mean difference of tibial SOS at 40 weeks was comparable. DSOS from baseline to 40 weeks was also not different between the groups (Fig. 2). There was no difference in weight, length or head circumference between the two groups at 40 weeks. Serum calcium, phosphorus, and alkaline phosphatase levels at 40 weeks were also similar in the two groups (Table III).

TABLE III Anthropometric and Biochemical Measurements at 40 Weeks

Post hoc, sub-group analysis was done to compare infants in each strata with GA 27^0/7 to 32^6/7 and 330/7 to 34^6/7 respectively, and also by their intrauterine growth restriction (IUGR) status (Web Table I). There was no difference noted between the groups in tibial bone SOS at 40 weeks.

1. Forbes GB. Calcium accumulated by the human fetus. Pediatrics.1976;57:976-7.

2. Ziegler EE, O’Donnell AM, Nelson SE, Fomon SJ. Body composition of the reference fetus. Growth. 1976;40: 329-41.

3. Callenbach JC, Sheehan MB, Abramson SJ, Hall RT. Etiologic factors in rickets of very low birth-weight infants. J Pediatr. 1981;98:800-5.

4. James JR, Condon PJ, Truscott J, Horsman A, Arthur R. Osteopenia of prematurity. Arch Dis Child. 1986;61: 871-6.

5. Blackstrom MC, Kuusela AL, Maki R. Metabolic bone disease of prematurity. Ann Med. 1996;28:275-82.

6. Mazess RB, Whedon GD. Immobilization and bone. CalcTiss Res. 1983;35:265-7.

7. Schulzke SM, Kaempfen S, Trachsel D, Patole SK. Physical activity programs for promoting bone mineralization and growth in preterm infants. Cochrane Database Syst Rev. 2014;4:CD005387.

8. Ballard JL, Khoury JC, Wedig K, Wang L, Eilers-Walsman BL, Lipp R. New ballard score, expanded to include extremely premature infants. J Pediatr. 1991;119:417-23.

9. Moyer-Mileur L, Luetkemeier M, Boomer L, Chan GM. Effect of physical activity on bone mineralization in premature infants. J Pediatr. 1995;127:620-5.

10. Agarwal R, Deorari A, Paul VK. AIIMS Protocols in Neonatology. New Delhi: CBS Publishers & Distributors Pvt Ltd, 2015.

11. Natarajan CK, Sankar MJ, Agarwal R, Pratap OT, Jain V, Gupta N, et al. Trial of daily vitamin D supplementation in preterm infants. Pediatrics. 2014;133:e628-34.

12. Litmanovitz I, Dolfin T, Friedland O, Arnon S, Regev R, Shainkin-Kestenbaum R, et al. Early physical activity intervention prevents decrease of bone strength in very low birth weight infants. Pediatrics. 2003;112:15-9.

13. Litmanovitz I, Dolfin T, Arnon S, Regev RH, Nemet D, Eliakim A. Assisted exercise and bone strength in preterm infants. Calcif Tissue Int. 2007;80:39-43.

14. Tosun O, Bayat M, Gunes T, Erdem E. Daily physical activity in low risk preterm infants: Positive impact on bone strength and mid – upper arm circumference. Ann Hum Biol. 2011;38:635-9.

15. Erdem E, Tosun O, Bayat M, Korkmaz Z, Halis H, Gunes T. Daily physical activity in low-risk extremely low birth weight preterm infants: positive impact on bone mineral density and anthropometric measurements. J Bone Miner Metab. 2015;33:329-34.

16. Moyer-Mileur LJ, Bail SD, McNaught TP, Chan GM. Effect of physical activity on bone mineralization and body composition in preterm infants during the first year of life. Pediatr Res. 2000;47:292A.

17. Litmanovitz I, Erez H, Eliakim A, BaurRusek S, Arnon S, Regev RH, et al. The effect of assisted exercise frequency on bone strength in very low birth weight preterm infants: a randomized control trial. Calcif Tissue Int. 2016;99:237-42.

18. Eliakim A, Nemet D, Friedland O, Dolfin T, Regev RH. Spontaneous activity in premature infants affects bone strength. J Perinatol. 2002;22:650-2.

19. Moyer-Mileur LJ, Brunstetter V, McNaught TP, Gill G, Chan GM. Daily physical activity program increases bone mineralization and growth in preterm very low birth weight infants. Pediatrics. 2000;106:1088-92.

20. Nemet D, Dolfin T, Litmanowitz I, Shainkin-Kestenbaum R, Lis M, Eliakim A. Evidence for exercise-induced bone formation in premature infants. Int J Sports Med. 2002;23:82-5.

21. Vignochi CM, Miura E, Canani LH. Effects of motor physical therapy on bone mineralization in premature infants: a randomized controlled study. J Perinatol. 2008;28:624-31.