Safety and efficacy of double plastic wrap in reducing insensible water loss in preterm infants in first week of life – a pilot randomized controlled trial from a low-to-middle-income country

Abstract

Background

Preterm infants experience excessive insensible water loss in postnatal period. Established practices like cling film might not be sufficient alone in reducing this loss. Being expensive, humidified incubators might not be affordable in developing countries. Hence, we tried to explore double plastic wrap (cling film attached to the bassinet wall and occlusive plastic wrap covering the baby) as a low-cost solution of increased insensible water loss in postnatal period.

Methods

In this pilot trial 63 inborn infants of less than 32 weeks of gestation or birth weight less than 1200 g were enrolled and randomized to either single wrap (only cling film attached to the bassinet wall) or double wrap (cling film covering the bassinet and occlusive plastic wrap covering the baby) group after birth. This practice was allowed in addition to radiant warmer care and other routine thermoregulatory measures followed in the unit. Intervention was continued till first 7 days of life unless they met any pre-specified withdrawal criteria. The primary outcome was a difference in total insensible water loss (g/m2) in first seven days of life between two groups. The secondary outcomes were difference in cumulative insensible water loss (g/m2) in the first 72 h of life, 4–7 days of life, average daily insensible water loss in the first week (g/m2/h), neonatal morbidities, mortality and time to event analysis.

Results

Of 63 randomized infants, 32 were allocated to single wrap and 31 to double wrap groups. Cumulative insensible water loss in first 72 h and first week were significantly lower in the double wrap group (2786.5 ± 576 g/m2 in single wrap and 2376.8 ± 626 g/m2 in double wrap, p value 0.012 for 0–3 days and 6225.48 ± 951 g/m2 in single wrap and 5260.61 ± 1091 in double wrap, p value 0.034 for 0–7 days of life respectively). Apart from increased incidence of patent ductus arteriosus in double wrap group (41.94% vs. 15.62%, p 0.027) no other clinically significant outcomes were different between 2 groups. Median times to wean off respiratory support, regain birth weight and discharge from health facility were also similar between the two groups.

Conclusion

Double plastic wrap has been found to reduce cumulative insensible water loss in first week of life compared to cling film alone in the setting of a low-to-middle income-country.

Trial registration number

CTRI/2024/03/063749.

Trial registration date

07.03.2024.

Link to CTRI

https://ctri.nic.in/Clinicaltrials/pmaindet2.php?EncHid=MTAwMzk5&Enc=&userName=CTRI/2024/03/063749.

Background

Fluid and electrolyte homeostasis is a challenge in newborns, especially in preterm. Owing to immature epithelial layer and larger surface area, trans epidermal water loss is markedly high in them contributing to increased morbidity and mortality [1]. Additionally due to requirement of respiratory support in these preterm newborns, insensible water loss (IWL) through respiratory tract also contributes to the burden. To overcome increased IWL in these preterm newborns, various strategies have been adopted such as plastic bags, thermostable gel mattresses and high-humidity environments like incubators [2] So far measures like occlusive plastic wraps or bags have been applied to newborns at delivery point and during transport from delivery room to neonatal intensive care unit (NICU) [3]. Application of occlusive wrap in very low birth weight infants at delivery point improves admission temperature contributing to a decreased mortality rate in them [4]. Occlusive wrap must be made of polyethylene rather than polystyrene because only polyethylene transmits the long wavelength energy of radiant heat [5]. However, there is scarce evidence on the continuation of such measures in post-natal setting in NICU.

In an ideal postnatal setting, preterm infants should be kept in a thermo-neutral environment inside a servo-controlled incubator with high humidity in the first few weeks of life to prevent insensible water loss. Cochrane studies have reported that incubators are effective in reducing insensible water loss and improving electrolyte balance in preterm newborns due to a highly humidified environment [6] However, in low- and middle-income countries (LMIC) like India, incubators might not be a feasible option due to high cost, risk of sepsis and also the lack of expertise required for managing and maintaining the infant inside [7]. A Cochrane meta- analysis by Flenady VJ et al. showed that radiant warmers result in increased IWL compared to incubators (mean weighted difference of 0.94 g/Kg/day (95% CI 0.47, 1.41)) [8]. Cling film is a simple and biodegradable food grade plastic covering over the warmer bassinet, which creates a micro-environment of increased humidity and limited air movement. So, this serves as a safe, cost-effective and affordable intervention to reduce insensible water loss in preterm newborns in the setting of a developing country [9, 10]. However, there is still significantly more IWL compared to incubator (mean difference in IWL in ml/kg/h: 1.24[95% CI 0.3, 2.18]) [8]. An alternative low-cost solution could be keeping the baby under double plastic wrap (cling film attached to the bassinet walls and occlusive wrap covering the baby) while nursing the baby in warmer. With this idea we have studied the feasibility, safety and efficacy of double plastic wrap in reducing insensible water loss in preterm newborns when nursed under radiant warmer in the first week of life in the setting of a LMIC.

Methods

An open label parallel arm pilot randomised controlled trial (RCT) was conducted in a tertiary care neonatal unit of eastern India from March to August 2024. The study was conducted in line with the principles of declaration of Helsinki [11]. Manuscript was written in line with the CONSORT guideline for pilot and feasibility trials. All inborn infants born at less than 32 weeks of gestation or with birth weight less than 1200 g during the study period were enrolled. Infants with major congenital anomalies (any internal or external structural defect identifiable at birth requiring significant surgical or cosmetic intervention), suspected chromosomal anomalies, foetal hydrops, 20 min APGAR score ≤ 3 and parental refusal of consent were excluded from the study.

Parents were counselled before birth regarding the importance of maintaining thermal comfort for the baby in postnatal life. Gestational age was checked before birth by first trimester dating scan and date of last menstrual period (LMP). Best obstetric estimate of gestational age was taken (Dating scan corroborated with LMP > dating scan alone > LMP alone) and noted in the case record sheet. If the baby was anticipated to be < 32 weeks, a neonatologist fellow attended the delivery. Baby was received in a polyethylene occlusive plastic wrap (Neo-HeLP by VYGON of small and medium sizes with dimensions of 30*38 cm and 38*44 cm respectively depending on the birth weight). Latest neonatal resuscitation protocol (NRP 2020) was followed for resuscitation including practise of delayed cord clamping (DCC), delivery room continuous positive airway pressure (DR CPAP), and skin to skin contact (SSC) as applicable. After ensuring eligibility criteria at birth parents were approached for written informed consent. If refused, baby was not enrolled in the study.

Eligible newborns were shifted to NICU wrapped in plastic bags without drying from delivery point of care. In NICU plastic wraps were removed to lower the transmission risk of infection from the place of delivery and placed under radiant warmer with a cling-film applied to the bassinet sidewalls. Radiant warmer was set at servo control mode with set temp of 36.5⁰C. Babies of consenting parents were randomly allocated into control and intervention arm after NICU admission. In the intervention arm, newborns were kept in a double wrap (occlusive plastic wrap covering the infant and cling-film attached to the side walls of bassinet) (Fig. 1) for first 7 days of life and in the control arm, newborns were kept in single wrap (only cling film covering over the warmer bassinet) (Fig. 2) for the same duration while being nursed under radiant warmer.

Fig. 1

Newborn in polyethylene occlusive plastic wrap

Fig. 2

Newborn in single wrap (plastic wrap over warmer bassinet)

Both intervention and control groups received standard care as part of unit policy which included controlled NICU environment, heated humidified gases during ventilation, coconut oil application and cling wrap. Total fluid intake (maintenance fluids and medications) was initiated at a rate of 80 ml/kg/day on day 1 and increased up to 150 ml/kg/day on day 7, thereby incrementing at a rate of 15–20 ml/kg/day. The fluid intake of each newborn was adjusted per day depending on various factors like serum sodium levels, body weight, and urine output. Electrolytes were added to the maintenance fluids at day 2–3 of life and decision of parenteral nutrition was left to the treating physician.

Block randomization was done using variable block sizes of 4, 6, 8 (www.sealedenvelop.com) by a separate person not involved in the study. Allocation concealment was done using serially numbered opaque sealed envelope (SNOSE) which were kept at the nursing station and only opened after randomization by the treating physician. Information about demographic profile, perinatal risk factors including maternal obstetric and systemic complications, birth history, resuscitation details, daily change in body weight, fluid intake and output, serum sodium, respiratory support, systemic morbidities and final outcome were recorded in a predesigned proforma.

The primary outcome of this study was to compare cumulative IWL in 2 arms (double vs. single wrap) in the first seven days of life. Body surface area (BSA) was calculated by formula given by Mosteller [√ (height in cm * weight in kg)/3600] [12]. Density of commonly administered fluids and drugs (Table 1) were calculated by dividing the weight difference of full and empty container with the respective drug/fluid volume. Daily IWL was calculated using the following equation: IWL (gram/m2/hour) = [(Total fluid intake (gm) – Urine output (gm)) ± weight difference from next day (in grams)]/body surface area (m2)/24 [1]. The infant’s weight was recorded on digital weighing scale with accuracy of ± 5 g at the beginning of each morning shift. The final weight was calculated after subtracting premeasured weights of ancillary objects like endotracheal tube (5 g), canula (3 g), diaper (20 g). Weight difference from next day is added if there is weight loss and subtracted if there is weight gain. Volume of the total intake was calculated by adding up all the inputs (fluid, caffeine, drugs etc.) and converted into grams by multiplying with the corresponding densities. Water lost in stool was not considered. Cumulative IWL (g/m2) was calculated by summing up total daily IWL for a given period. Total IWL was calculated two time periods- IWL for first 72 h of life (day0-day 3) and beyond 72 h of life (day 4 to day 7).

Table 1 Density of commonly administered fluids and drugs

Secondary outcomes were daily average insensible water loss(g/m2/h) in first week of life, requirement of maximum respiratory support, morbidities like sepsis (clinical and culture positive), meningitis, neonatal skin score [13], moderate or severe hypothermia(axillary temperature < 36⁰C), hyperthermia(axillary temperature > 37.5⁰C), hypoglycaemia(serum glucose < 45 mg/dL), hyperglycaemia (serum glucose > 145 mg/dL), hyponatremia (serum sodium < 130 mEq/L), hypernatremia(serum sodium > 145 mEq/L), pulmonary haemorrhage, acute kidney injury (AKI), haemodynamically significant patent ductus arteriosus (HS PDA), grade 3/4 intra-ventricular haemorrhage(IVH), bronchopulmonary dysplasia (BPD), retinopathy of prematurity (ROP), definite or advanced necrotising enterocolitis (NEC), maximum postnatal weight loss, time to regain birth weight, duration of respiratory support, time to death and time to discharge from health facility. Neonatal skin condition was assessed by allotted nursing staff at each shift. Each baby was continuously monitored by servo-controlled probe attached to the warmer. Any reading below 36 ⁰C or above 37.5⁰C was immediately confirmed by digital axillary thermometer. Hypothermia or hyperthermia was defined by axillary temperature readings. BPD was defined by oxygen requirement of first 28 days of life. ROP was defined by ICROP classification. Serum electrolytes were monitored daily for first week of life. Capillary blood sugar (CBS) was tested every 8 hourly. Any CBS value below 50 was confirmed by serum blood glucose estimation. Hypoglycemia was defined by serum values. Echocardiography and neurosonography was done as per treating physician’s discretion.

Sample size

Several researchers have recommended various sample sizes for conducting a pilot RCT ranging from 20 to 70 [14,15,16]. In the absence of any consensus, we considered a sample size of 50. Those infants who expired within first week of life were not considered in calculating cumulative IWL from day 0 to day 7. Considering 20% death rate in first week of life a total sample size of 60 was estimated.

Interim analysis was performed at 50% completion of the estimated sample size and safety and efficacy of the intervention was analysed. Following adverse events were recorded during the trial–mortality (at the end of first week and at final outcome), weight gain by > 3% for 3 consecutive days, daily weight loss of > 3% for 3 consecutive days, cumulative weight loss of > 20%, severe hyponatremia (< 120 meq/L), severe hypernatremia (> 150 meq/L), severe hyperthermia (> 39⁰C) and severe hypothermia (< 32⁰C). Infant with any adverse event was withdrawn from the study and treated accordingly. As the proportion of infants < 28 weeks were small, we did not consider hsPDA as a withdrawal criterion from the study. Adverse events and cumulative IWL in first week were compared between both groups (single vs. double wrap) at interim analysis and a p value of less than 0.025 would lead to termination of the trial [17].

Statistical methods

Categorical variables were described as frequencies and percentages. Continuous variables were assessed by histogram and QQ plot to check for normality. Normally distributed continuous variables were described by mean (± standard deviation) and skewed distribution by median (inter quartile range) respectively. Continuous outcomes (e.g. daily IWL, total IWL etc.) were analysed by student’s t test (for parametric data) or Mann-Whitney U test (for nonparametric data). Categorical outcomes (e.g. - incidence of BPD, PDA, mortality etc.) were tested using Fisher’s exact test. A survival analysis using Kaplan –Meier method and log rank test was done to determine any difference in time to event analysis between both groups. Analysis was performed by intention to treat (ITT) principle. A significance level of 0.05 was considered as a priori. All statistical analysis was done using R software (version 4.3.2).

Results

During the study period 73 infants were assessed for eligibility of which 63 were randomized into 2 groups (32 infants in single wrap group and 31 infants in double wrap group). The CONSORT flow diagram is shown in Fig. 3. Baseline characteristics of the study population are shown in Table 2. Both groups received allotted intervention for comparable time period. 92.06% infants had records for first 72 h of life whereas 84.12% had records for day 4- day 7 of life. Reasons for missing values in each group are detailed in the CONSORT flow diagram (Fig. 3).

Fig. 3

CONSORT flow diagram. Legend: IWL – insensible water loss

Table 2 Baseline characteristics of the study population

Actual fluid received was similar between 2 groups, which has been shown in Table 3. Cumulative insensible water loss (g/m2) for first 72 h of life was significantly lower in the double wrap group (2786.5 ± 576 g/m2 in single wrap and 2376.8 ± 626 g/m2 in double wrap, p 0.012) for those who survived for the first 3 days and (6225.48 ± 951 g/m2 in single wrap and 5620.61 ± 1091 g/m2 in double wrap, p 0.034) for the first 7 days of life (Fig. 4 ). However average daily insensible water loss in first week was not different between 2 groups (37.7 ± 8.97 g/m2/h in single wrap vs. 33.2 ± 10.2 g/m2/h in double wrap, p 0.066) for all infants irrespective of their time of outcome. Daily rate of insensible water loss (g/m2/h) between 2 groups is shown in Fig. 5. Double wrap was effective in reducing daily IWL (g/m2/h) for first 2 days of life (95% CI mean difference ranges from − 10.77 to -3.22, p value 0.001 on day 1 and − 14.38 to -1.64, p value 0.014 on day 2). Daily IWL was not different in remaining days of the week.

Table 3 Actual day wise fluid intake (ml/kg)

Fig. 4

Cumulative insensible water loss (g/m2) in first week of life. Legend: IWL- insensible water loss

Fig. 5

Daily insensible water loss (g/m2/h) in first week of life. Legend: IWL – Insensible water loss, values are in g/m2/h. Black dots represent values in median

In first week of life there was no difference in surfactant requirement, body temperature, serum sodium, serum glucose, skin scoring or mortality between 2 arms (Table 4). However, incidence of HSPDA was significantly higher in the double wrap group (41.94% vs. 15.62%, p 0.027). There was no difference in other clinically significant outcomes including mortality (Table 5).

Table 4 Outcomes in first week of life

Table 5 Outcomes till hospital discharge

Median time to regain birth weight, duration of respiratory support, time to discharge was not different between 2 groups (Table 6). There were 6 newborns (9.52%) in our study born at less than 28 weeks of gestation. Higher incidence of co-morbidities like pulmonary haemorrhage, hsPDA and BPD could have contributed to prolonged duration of respiratory support in our study population. Median time to death in single wrap group was earlier than double wrap group although not statistically significant (median (95%CI):3 days vs. 9 (6, 9.75) days, p 0.1).

Table 6 Time to event analysis

We examined any possible interaction effect between gestational age and the intervention which revealed average daily IWL in first week of life decreased by 3.6 g/m²/h in double wrap group for 1 week increase in gestational age at birth whereas average daily IWL decreased by 0.5 g/m²/h in the single wrap group for the same increment in gestational age, p 0.022) (Fig. 6). This suggests that double wrap becomes more efficient from 28 weeks onwards as it significantly reduces average daily IWL compared to clingfilm alone beyond this point. Infants of less than 28 weeks experience greater IWL compared to more mature infants because of lack of subcutaneous fat, immature epithelial barrier, increased skin permeability and plastic wrap (be it single or double) might not be sufficient enough for reducing it.

Fig. 6

Interaction plot between gestational age and intervention (occlusive plastic wrap)

Discussion

In this pilot study, double plastic wrap has been found to be effective in reducing cumulative IWL in first week of life. Apart from increased incidence of HS PDA no other clinically important side effects were noticed.

As it was a novel idea, we also checked feasibility of the intervention. We found it difficult to maintain in babies with chest drain and had to withdraw one infant from the trial. Apart from this, double plastic wrap was easy to use and maintain even in resource limited setting with limited patient to nurse ratio. However, daily change of plastic wrap and monitoring of skin condition at each shift was done as a part of the study protocol.

Contribution of evaporative heat loss is highest immediately after birth which gradually decreases with advancing post-natal age [18]. It becomes non-significant beyond first week of life [19]. Hence, we considered IWL in first week of life as our outcome of interest. Double wrap decreased cumulative IWL in first 3 days of life in those who survived that period. As IWL decreases with advancing post-natal age double wrap did not have same efficacy beyond 72 h of life. Because of its effect in first 3 days, total IWL in first 7 days were also low in the double wrap group. However, we did not find any difference in average daily IWL in first week as average tends to mask any small difference between groups and it would require a large sample size to find any significant difference in average IWL between 2 groups [20]. Increased fluid retention in the double wrap group might explain increased incidence of PDA. Although not statistically significant, incidence of AKI and pulmonary haemorrhage was more in the double wrap group. It could be explained by increased incidence of PDA in the double wrap group.

With decrease in gestational age, the trans epidermal water loss is increased, reaching 50 g/m2 /h at 26 weeks [21]. However, in this study we found a relatively lower rate of average IWL in first week of life (37.71 ± 8.74 g/m²/hour) even in the single wrap group. This could be explained by relatively mature infants (9.52% of newborns being born at less than 28 weeks of gestation), use of heated humidified gases during ventilation (37.5 degree C at 100% relative humidity), controlled thermal environment of NICU (26–28 degree C temperature and 40–45% humidity), oil application, and continued use of cling wraps for all newborns in the first week of life in our setting.

Infants who died in the single wrap group died early than those who died in the double wrap group (3 days vs. 9 days). Though it was not statistically significant, it was clinically important. However, being a pilot study it was underpowered to detect any survival difference between 2 intervention arms.

Lastly, we noticed an important interaction effect of gestational age with the intervention. There was a decrease in average daily IWL in first week with increasing gestation age at birth. However, this drop was quite marked in the double wrap group suggesting it might be more effective with increasing gestational age. In this study double wrap was found to be most effective from 28 weeks onwards.

The strength of this study lies in its robust methodology, timely recruitment, detailed analysis and inclusion of population representative of LMIC. This study highlighted markedly different conditions in a LMIC such as low coverage of antenatal steroid (19%), greater incidence of intrauterine growth retardation (31.74%), low rate of delayed cord clamping (38%) or use of CPAP at delivery point (66.67%). Large number of unbooked mothers and late referral from periphery for delivery at our center could have contributed to low coverage of antenatal steroids and relatively higher incidence of intrauterine growth retardation. High patient load, lack of manpower, lack of resources, and knowledge- practice gap might have contributed to low rate of DCC and/or CPAP. The novelty of this pilot study lies in the fact that we tried to explore the safety and efficacy of a low-cost intervention like double plastic wrap in the setting of a LMIC. Double plastic wrap was more effective than single wrap for reducing insensible water loss in the first 3 days of life, although there was an increased risk of hsPDA.

This study has several limitations as well. Being a pilot study, proper sample size calculation was not performed. Blinding was technically not feasible in this study. Approximately 12.7% (8 out of 63) newborns died within first week of life contributing significant loss of outcome data. In this study insensible water loss was calculated not directly measured. Researchers have found difference between measured and estimated insensible water loss values in newborn infants [22]. Incidences of certain outcomes, such as pulmonary hemorrhage, AKI were higher in double wrap group, however it did not reach statistical significance. This should be explored further in future trials. As most of the infants in our study population were sick, we could not initiate kangaroo mother care (KMC) in first week of life until they achieved hemodynamic stability. Hence, impact of KMC on the intervention could not be assessed. Future studies should explore the safety and efficacy of double plastic wrap in adequately powered randomized controlled trials in resource limited settings.

Conclusion

In the absence of incubators, use of double plastic wrap along with other thermoregulatory measures in preterm newborns can lead to reduced cumulative insensible water loss in first week of life at the cost of increased incidence of PDA. However, it does not translate into improved clinical outcomes.