Effect of osteopathic manipulative treatment on length of stay in a population of preterm infants: a randomized controlled trial

Preterm birth is defined as an early delivery made before the 37th week of gestational age [1–3]. The burden of preterm newborns has been steadily rising worldwide during the last 20 years [3]. In 2005 the global rate of preterm newborns was 9.6% (95% C.I. 9.1–10.1) per 12.6 million births per year [3]. This rate differs across regions, with the highest proportion of preterm births in Africa (11.9%; 11.1–12.6) and the lowest in Europe (6.2%; 5.8–6.7) [3].

Preterm birth leads to an increased lifetime morbidity and healthcare costs due to higher rates of neurological, respiratory, cardiovascular and psychological disorders [4–6].

With regards to economic costs, Russell et al. (2007) estimated the annual cost of preterm-birth on NICU to be 5.8 billion dollars (US), corresponding to 47% of the overall cost for all infants hospitalization and to 27% of all pediatric stays [7]. Data from another study carried out in the UK on preterm infants born at 30 weeks, documented mean costs per length of stay (LOS) for premature infant being higher than £10,000 (~$15,500) compared to £1300 (~$2080) for term infant [8, 9].

LOS seems also to be strongly related to the infant’s physiological status [10], in terms of cardiorespiratory functions, full enteral feeding (bottle and/or nipple) and stooling. These preterms’ growth indicators are clinically relevant, in term of achievements that must be reached before they are safely discharged home [10]. The majority of these physiological milestones are achieved between 34 and 36 weeks of post-menstrual age, although these are secondary to the patient’s clinical condition [10]. The same authors also pointed out that feeding and oxygen goals are usually achieved last compared to all others. As a consequence, all of these factors have a significant influence on LOS. Therefore, taking into account the above aspects of the care of preterm infant, LOS may be considered an important outcome in relation to both the clinical status of the patient and the economic impact of services provided by NICUs.

As far as complementary and alternative medicines are concerned, few studies have been conducted in the preterm infant population. A review on massage therapy showed that the application of a 10-15 min massage led to an improvement of weight gain and to a LOS reduction [11]. However, Cochrane’s review suggested that these findings were inconsistent and highlighted many methodological concerns with previous studies [12].

Furthermore, from an osteopathic medicine perspective, there are no randomized control trials on premature infants. In 2011 Pizzolorusso et al. carried out an observational study of N=352 infants, suggesting that osteopathic manipulative treatment (OMT) may play an important role in reducing the risks for LOS longer than 28 days (adjusted OR=0.45;0.26-0.74) and the number of episodes of vomit, regurgitation, gastric residual and enema (adjusted OR=0.22;0.09-0.51) [13].

The aim of this study is to confirm and investigate, using gold standard methods, the effectiveness of OMT in LOS in a population of preterm infants.

Variables considered as possible confounders that were included into the model are: gender, gestational age, weight at birth, milk volume at study enrollment and OMT.

During the trial, all subjects received banked milk and 11 preterms (5 study group and 6 control group, X ²=0.04, p=0.84) were additionally fed with breast milk.

At the end of the study, 9 subjects dropped out due to hospital relocation (Figure 1). Reasons were: confirmation of genetic disorders (3 subjects from OMT group), bacterial infection complications (2 subjects from OMT group), cerebral hemorrhage (1 subject from the OMT group and 1 from the control group), convulsion (1 subject from OMT group) and cerebral hypoxia (1 subject from OMT group).

The mean LOS average was 26.1±16.4 for the study group whilst 31.3±20.2 for the control group (p<0.03).

A generalized linear model was performed to detect the real difference in mean between the two groups, taking into account confounders for the main outcomes.

Results for LOS are shown in Table 2 and Figure 2. All variables were found to be associated with a change in LOS, except for gender and milk volume at study enrollment.

	LOS (days)			Av. daily weight gain (gr)
	β	95% C.I.	Pr(>|t|)	β	95% C.I.	Pr(>|t|)
Male	-1.899	-3.930 , 0.127	0.07	0.708	-3.067 , 4.483	0.71
Gestational Age	-3.373	-3.916 , 2.830	<0.001	-0.338	-1.344 , 0.668	0.51
Birth Weight (gr)	-0.014	-0.016 , -0.009	<0.001	-0.018	-0.022 , -0.014	<0.001
Milk Volume at Study Enrollment (mL)	0.002	-0.004 , 0.009	0.44	0.059	0.045 , 0.072	<0.001
OMT	-5.906	-7.944 , -3.869	<0.001	3.707	-0.065 , 7.479	0.06

A negative association was found for gestational age (-3.373; 95% C.I. -3.916, - 2.830; p<0.001), birth weight (-0.014; -0.016, -0.009; p<0.001) and OMT (-5.906; -7.944, -3.869; p<0.001).

Results for weight gain are shown in Table 2 and Figure 3. Only birth weight (-0.018; -0.022, -0.014; p<0.001) and milk volume at admission (0.059; 0.045, 0.072; p<0.001) were found to be associated with changes in weight gain, while gender, gestational age and OMT were not statistically significant associated.

No adverse and side effects were shown in both groups during the study period.

The results from the present study highlight the potential benefit of OMT in newborns in our population. LOS reduction in the OMT group was highly significant compared to the standard care group (26.1 vs 31.3). This indicates that the OMT can shorten the period of hospitalization of almost 6 days, opening up several considerations from a health care prospective.

LOS has been widely used as an outcome measure in research contexts. Moreover, it has been considered a valid proxy measure to evaluate the clinical status in pre-term infants [10]. If the correlation between LOS and infant physiological status is considered to be strong [10], then these results would indicate that OMT in pre-term infants leads to significant benefits such as cost reductions, enhanced quality of health care service delivered and improved health status of pre-term infants.

Few studies are available in the medical literature in relation to the pediatric population and no structured OMT clinical trials have been carried out in preterm infants during their stay in NICUs.

Some data are available from studies on a wider population of infants and children, documenting the positive effects of osteopathic care in gastrointestinal function [21], obstructive apnea [22] and postural asymmetry [23].

There has been only one recent prospective study carried out in Italy [13], where the authors looked at the effect of OMT on LOS and gastrointestinal functions. Pizzolorusso et al showed that preterm infants treated with OMT group significantly reduced the hospital LOS (adjusted OR=0.22;0.09-0.51). Moreover, they showed that infants in the intervention group had a significance reduction in the risk of excessive gastrointestinal symptoms, such as vomit, regurgitation, gastric residual and enema administration (adjusted OR=0.45;0.26-0.74). These findings lead to consider a possible and potential role of OMT in increasing the global health condition of newborns. However, the study enrolled and analyzed all newborns accessing at the NICU without discriminating between pre-term and term neonates.

The mechanisms of action of OMT are not fully understood. Hypotheses based on clinical data are as follows. The first hypothesis takes into account the anti-inflammatory action of OMT. Narendran et al (2010) showed that prematures have a higher level of cortisol, albumin, IL-8 and IL-1β [24], suggesting an increased level of systemic inflammation. A study carried out in 2007 demonstrated that OMT can reduce the inflammatory process acting mainly on anti-inflammatory factors [25]. Moreover, Degenhardt et al suggested that OMT could have a role in increasing the opioid reaction [26]. Therefore OMT may potentially modulate and reduce the inflammatory status of infants through the anti-inflammatory mechanism. However, this hypothesis has some intrinsic limitations in terms of the sample used, i.e. no infants were osteopathically treated, and in translating vitro findings into vivo mechanisms.

Another possible hypothesis involves the function of the autonomic nervous system (ANS). Longin et al [27] demonstrated that gestational age of newborn infants is correlated with a change in the heart rate variability (HRV). Furthermore prematures have a different level of HRV ratio leading to considerably different reactions of the ANS. From the osteopathic perspective, changes in the HRV were recorded after the application of myofascial release techniques [28], as demonstrated by Henley et al. in 2008. For this reason, the application of OMT could bring balance to the sympathetic and para-sympathetic inputs, creating an improvement of newborns clinical condition. As for the previous hypotheses, the different sample does not allow to predict a similar reaction in preterm infants.

According to costs computation, the application of OMT can theoretically save almost 3,000€ per preterm per LOS, putting the NICU in a more favorable economic position. The latter is due to the ISS reimbursement procedures which establish a payment method based on DRG mean cost estimates [18] rather than on the duration of a single hospital admission. The net saving could be considered a “moneybox” that NICU managers can use to improve efficiency in procedures, implementing diagnostic tools or optimizing turnovers. The cost analysis is highly dependent upon the relationship of the cost-reimbursement of the OMT to the cost-reimbursement of the DRG. Considering the cost for OMT session equal to the cost of a standard physical care in a public domain (25€) [18], the net saving is approximately 2,700€ per LOS per newborn, confirming the osteopathic treatment a valuable intervention.

As a matter of fact, in a decentralized national health care system, osteopathy in NICUs can produce more benefits to the single regional and local health care policymakers rather than to ISS payment system, following the national criteria to propel more effective health care methods.

Some limitations of the present study need to be outlined, in terms of small sample size and all patients enrolled are referred to one single NICU and may be not representative of the preterm infants population. Moreover maternal socioeconomic status was not collected and it might be another potential confounder. Furthermore, during the study period, hospital relocation was predominately from the study group. This could be related to potential adverse effects of OMT, the type of the population, the relative small sample size and the fragile newborns’ conditions can casually predispose to this difference. Additional studies with lager sample size are needed to confirm these results.

As far as the economic evaluation is concerned, the cost estimation was based on a theoretical approach based on mathematical computation of hospital net savings. The major limitation of this approach is due to the national and regional reimbursement system which based its activity on diagnosis category (DRG) with a theoretical LOS threshold rather than on the real length of stay [18]. This produces theoretical savings that might not be representative of the real hospital scenario.

The treatment of preterm infants with complementary and alternative medicine could be considered a potential and promising intervention to be included into the routine health care system. Results from the present study show the effectiveness of the OMT in reducing the length of stay, costs but not weight gain in a sample of pre-term newborns. Despite the lack of prospective research into the effects of osteopathy on preterm infants, our study shows strong evidence that OMT has a role in the care of newborns. Further studies based on multi-centric design are required to confirm our results.