Sniffin’ Away the Feeding Tube: The Influence of Olfactory Stimulation on Oral Food Intake in Newborns and Premature Infants

Sniffin’ Away the Feeding Tube: The Influence of Olfactory Stimulation on Oral Food Intake in... Abstract Because of their immaturity, many premature infants are fed via nasogastric tube. One objective of the neonatal care is to feed infants orally early. The olfactory function of premature infants is developed before birth and odorants have a significant impact on nutrition in infants. The aim of the study was to test whether odor stimulation has a positive effect on the transition from gavage to oral feeding in infants. Participants were premature infants with gestational age of more than 27 weeks, with full or partial gavage feeding, stable vital parameters and without invasive ventilation. Before each feeding procedure an odorant was presented in front of the infant’s nose. Infants were randomized into 1 of 3 groups and received either rose odor (not food-associated), vanilla odor (food-associated), or placebo (no odor). The primary outcome of the study was defined as the time until complete oral nutrition. About 150 children born at a postnatal age of 9.5 ± 7.8 days were included in this study. The duration until complete oral nutrition was reached after 11.8 ± 7.7 (vanilla), 12.2 ± 7.7 (rose), and 12.9 ± 8.8 (control) days. A nearly linear relation between odor presentation frequency and effect size was detectable. For infants that received the intervention for more than 66.7% of the time the length of gavage feeding (8 ± 5.4) and hospitalization (11 ± 6.5) was significantly lower in the vanilla group when compared with control (15 ± 7.3 and 21 ± 13.7, respectively). Odor stimulation with vanilla has an impact on oral feeding in premature infants, however the odor has to be presented on regular basis. gavage, nutrition, olfaction, premature infant Introduction The clinical and scientific progress in neonatology continuously improves the survival of premature infants especially with very low birth weight. Therefore, the focus is shifting to improve therapy and outcome. One major target is an adequate and early enteral feeding of premature infants. The majority of premature infants need to be fed by gavage due to insufficient sucking and a lack of suck–swallow–breath coordination (Gewolb and Vice 2006). As a result of gavage-feeding infants are not able to experience the flavor, a combination of smell and taste, of breast or formula milk. The physiological connection between the odor of nutriment, which initiates the cephalic phase, followed by sucking behavior and the satisfactory feeling of gastric filling is interrupted. Although gavage feeding is necessary and superior to parenteral nutrition (Calkins et al. 2014), the lack of olfactory perception during feeding might result in a prolonged transition from gavage to oral feeding. The importance of olfaction for newborns, especially of mother-associated odorants is well-known (Russell 1976; Varendi et al. 1994). Olfactory receptor neurons were found in preterm infants as early as 24–27 weeks of gestation (Chuah and Zheng 1987; Johnson et al. 1995). Therefore, it can be assumed that the olfactory sense is functioning in preterm infants. In 1994, a study showed the preference of infants for the odor of their mothers’ breast through head-movement towards the unwashed mother’s nipple (Varendi et al. 1994). Food-associated odorants are also important for the first days after birth. For example, 2 studies demonstrated an increase of non-nutritive sucking in response to milk odorant (Bingham et al. 2003, 2007). Non-nutritive-sucking is known to be important for infants to learn a coordinated suck–swallow–breathing procedure (Delaney and Arvedson 2008). Finally, a pilot study found a positive effect of breast milk odor on the duration of gavage feeding and hospitalization time of preterm infants born between gestation weeks 28–34 (Yildiz et al. 2011). Caused by the poor practicability of presenting fresh breastmilk odor in daily routine in an intermediate care unit, it was necessary to select another food-associated odor for this study. Stimulation with vanilla odor is not only food-associated, side effect-free and safe but has already been used in studies with infants showing an increase of sucking behavior (Mennella and Beauchamp 1996) and other positive effects, e.g. a reduction of apnea (Marlier et al. 2005). Based on previous literature, the question arises whether olfactory stimulation in general or only food associated odors support transition from gavage to full oral feeding in premature infants. Therefore, the study was designed using rose and vanilla odor to evaluate whether an intervention of odor stimulation before feeding compared to a control group leads to faster transition from gavage to complete oral feeding. Material and methods This prospective randomized controlled study was approved by the local Ethics Committee of the Medical Faculty at the TU Dresden (EK 356092014). The study was conducted in accordance with the Declaration of Helsinki on Biomedical Studies Involving Human Subjects. Participants Participants were premature infants born at a gestational age of more than 27 weeks with stable vital parameters and without invasive ventilation or CPAP at time of randomization. All newborns received complete or partial gavage feeding. Exclusion criteria were neurological disabilities, intracerebral hemorrhage >II°, confirmed diagnosis of a syndromal disease or anatomical gastrointestinal anomalies. They were recruited shortly after transition from intensive care to intermediate care unit. Written informed consent was obtained from the parents or legal guardians of all participating infants before inclusion in the study. Intervention Children were randomized into 1 of the following 3 study groups: (i) food-related odor “vanilla”, (ii) non-food associated odor “rose”, or (iii) an odorless probe (control). Using a predefined study list, controlling for gestational age children were randomized. Odors were presented to the infants before each feeding. The “Sniffin’ Sticks” were used for odor presentation (Kobal et al. 1996). For the odor presentation, the cap of the “Sniffin’ Sticks” was removed by the nurse and the tip of the pen was positioned for ~10 s ~2 cm under the nostrils of the infants. This procedure was repeated before each feeding by bottle or by nasogastric tube but not before breastfeeding in order to avoid any interference with the physiological mother–child interaction. After each odor presentation, the nurse documented the procedure in an electronic patient data management system (Integrated Care Manager, Dräger AG&co). Each infant received its own “Sniffin’ Stick” (Kobal et al. 1996). “Sniffin’ Sticks” are felt-tip pens, which are filled with an odorant instead of pigment (“Sniffin’ Sticks”, Burghart GmbH). The following substances were used as odors: Vanilla (4-Hdroxy-3-Methoxybenzaldehyd, Sigma&Aldrich #V1104) in solution with 1,2-Propanediol (Sigma&Aldrich, #134368; 3 g in 100 mL), Rose (Frey&Lau, #P0604034, diluted 1:10 in 1,2-Propanediol), the control group (placebo) received an odorless pen. Outcome The primary outcome of the study was defined as duration between study entry and complete oral food intake defined as solely oral feeding for at least 24 h. Body weight, amount of nutrition by gavage-, bottle- and breastfeeding was recorded daily until the end of hospitalization. Furthermore, a prospectively defined subanalysis was performed to determine the effect of percentage of odor presentation. Statistical analysis Analyses were performed using IBM Statistical Package for the Social Sciences version 23.0 (SPSS Inc.) software with significance set as P < 0.05. One-way ANOVA was used to test differences between the 3 study groups. Between-group comparisons were performed using Bonferroni-corrected posthoc tests. Whenever appropriate non-parametric tests were used for comparing not normally distributed samples. In order to determine the required minimal presentation time, the presentation time dependent course of effect size was calculated. This was done by the following routine: those 16 participants who received the odor stimulation for most time per odor were first averaged and the effect size was determined between baseline and outcome measurement. Sixteen participants were chosen in order to ensure a robust sample size for effect size calculation. For rose odor, this compared the presentation time from 63 to 100%; for vanilla odor this compared the time between 71 and 100 %. In the next step, we consecutively calculated the effect size for the next lower odor presentation time, hence 17 participants were included into the analysis. Thereafter, we included 128 participants and so on until the final effect size calculation comprised the whole sample. The minimal presentation time was plotted against the effect size per odor. A statistical calculation determining the necessary sample size was done a priory. For the comparison of the effect of odor stimulation on the transition from gavage to oral feeding between 3 groups an ANOVA should be applied. To obtain valid results with an effect size of f = 0.25 at an alpha level of 0.05 and a power of ß = 0.75 a study population of 141 infants would be necessary. Therefore, we decided to include 50 infants in each of the 3 study groups. Results A total of 150 healthy infants participated in the study, 15 children had to be excluded from the analysis for the following reasons: withdrawal (n = 3), transferal to other hospitals (n = 12). No infant had to be excluded due to adverse effects of the odor stimulation. The details of the baseline and birth data of the infants are presented in detail in Table 1. Table 1. Baseline characteristics of the 3 study groups: rose, vanilla, and placebo, n amount of children in the studygroup, mean ± SD Rose Vanilla Placebo Total Weight at birth in g 1862.5 ± 448.63 n = 46 1866.5 ± 424.6 n = 49 1889.9 ± 468.4 n = 40 1872.1 ± 442.9 n = 135 Length at birth in cm 43 ± 3.29 n = 46 43.3 ± 2.99 n = 49 42.9 ± 3.58 n = 40 43.1 ± 3.26 n = 135 Gestational week in days 231 ± 16.9 n = 46 230 ± 16.2 n = 49 230 ± 16.1 n = 40 230 ± 16.3 n = 135 APGAR 5 min 7.8 ± 1.12 n = 46 8.3 ± 0.93 n = 48 8.2 ± 1.15 n = 38 8.1 ± 1.08 n = 132 Age at randomization in days 9.8 ± 8.66 n = 46 9.2 ± 6.29 n = 49 9.4 ± 8.62 n = 40 9.5 ± 7.81 n = 135 Weight at randomization in g 1956.6 ± 411.76 n = 46 1916.5 ± 448.09 n = 48 1953.5 ± 384.98 n = 40 1941.3 ± 414.88 n = 134 total of gavage fed infants at randomization 2 2 2 6 % oral feeding at randomization 25.0 ± 20.78 n = 46 25.9 ± 21.0 n = 49 24.7 ± 18.69 n = 39 25.3 ± 20.13 n = 134 Rose Vanilla Placebo Total Weight at birth in g 1862.5 ± 448.63 n = 46 1866.5 ± 424.6 n = 49 1889.9 ± 468.4 n = 40 1872.1 ± 442.9 n = 135 Length at birth in cm 43 ± 3.29 n = 46 43.3 ± 2.99 n = 49 42.9 ± 3.58 n = 40 43.1 ± 3.26 n = 135 Gestational week in days 231 ± 16.9 n = 46 230 ± 16.2 n = 49 230 ± 16.1 n = 40 230 ± 16.3 n = 135 APGAR 5 min 7.8 ± 1.12 n = 46 8.3 ± 0.93 n = 48 8.2 ± 1.15 n = 38 8.1 ± 1.08 n = 132 Age at randomization in days 9.8 ± 8.66 n = 46 9.2 ± 6.29 n = 49 9.4 ± 8.62 n = 40 9.5 ± 7.81 n = 135 Weight at randomization in g 1956.6 ± 411.76 n = 46 1916.5 ± 448.09 n = 48 1953.5 ± 384.98 n = 40 1941.3 ± 414.88 n = 134 total of gavage fed infants at randomization 2 2 2 6 % oral feeding at randomization 25.0 ± 20.78 n = 46 25.9 ± 21.0 n = 49 24.7 ± 18.69 n = 39 25.3 ± 20.13 n = 134 View Large Table 1. Baseline characteristics of the 3 study groups: rose, vanilla, and placebo, n amount of children in the studygroup, mean ± SD Rose Vanilla Placebo Total Weight at birth in g 1862.5 ± 448.63 n = 46 1866.5 ± 424.6 n = 49 1889.9 ± 468.4 n = 40 1872.1 ± 442.9 n = 135 Length at birth in cm 43 ± 3.29 n = 46 43.3 ± 2.99 n = 49 42.9 ± 3.58 n = 40 43.1 ± 3.26 n = 135 Gestational week in days 231 ± 16.9 n = 46 230 ± 16.2 n = 49 230 ± 16.1 n = 40 230 ± 16.3 n = 135 APGAR 5 min 7.8 ± 1.12 n = 46 8.3 ± 0.93 n = 48 8.2 ± 1.15 n = 38 8.1 ± 1.08 n = 132 Age at randomization in days 9.8 ± 8.66 n = 46 9.2 ± 6.29 n = 49 9.4 ± 8.62 n = 40 9.5 ± 7.81 n = 135 Weight at randomization in g 1956.6 ± 411.76 n = 46 1916.5 ± 448.09 n = 48 1953.5 ± 384.98 n = 40 1941.3 ± 414.88 n = 134 total of gavage fed infants at randomization 2 2 2 6 % oral feeding at randomization 25.0 ± 20.78 n = 46 25.9 ± 21.0 n = 49 24.7 ± 18.69 n = 39 25.3 ± 20.13 n = 134 Rose Vanilla Placebo Total Weight at birth in g 1862.5 ± 448.63 n = 46 1866.5 ± 424.6 n = 49 1889.9 ± 468.4 n = 40 1872.1 ± 442.9 n = 135 Length at birth in cm 43 ± 3.29 n = 46 43.3 ± 2.99 n = 49 42.9 ± 3.58 n = 40 43.1 ± 3.26 n = 135 Gestational week in days 231 ± 16.9 n = 46 230 ± 16.2 n = 49 230 ± 16.1 n = 40 230 ± 16.3 n = 135 APGAR 5 min 7.8 ± 1.12 n = 46 8.3 ± 0.93 n = 48 8.2 ± 1.15 n = 38 8.1 ± 1.08 n = 132 Age at randomization in days 9.8 ± 8.66 n = 46 9.2 ± 6.29 n = 49 9.4 ± 8.62 n = 40 9.5 ± 7.81 n = 135 Weight at randomization in g 1956.6 ± 411.76 n = 46 1916.5 ± 448.09 n = 48 1953.5 ± 384.98 n = 40 1941.3 ± 414.88 n = 134 total of gavage fed infants at randomization 2 2 2 6 % oral feeding at randomization 25.0 ± 20.78 n = 46 25.9 ± 21.0 n = 49 24.7 ± 18.69 n = 39 25.3 ± 20.13 n = 134 View Large First, we tested whether there was a significant effect of group of presentation and whether there was an interaction between odor presentation time and group. We therefore conducted a univariate ANOVA (depend variable “time until total oral nutrition after randomization in days”) with the fixed effect of group and the covariate of odor presentation frequency and modeled the main effect of group and the interaction group by time until total oral nutrition after randomization in days. In results, there was a trend for group (F(2,120) = 2.469, P = 0.089) and a significant interaction effect (F(1,120) = 3.921, P = 0.01). In order to explore the interaction further, the results were analyzed more in detail. At baseline, the 3 study groups did not show a significant difference (Table 1). The duration between study entry and full oral feeding was 11.8 ± 7.7, 12.2 ± 7.7, and 12.8 ± 8.8 days for the vanilla, rose, and control group, respectively (Figure 1). Thus, the time of presentation was not significantly different between the 3 groups (F(2,123) = 0.24, P = 0.79). Taking a look at the feeding method (breast milk, formula, partial formula) at discharge from the hospital the feeding methods were equally distributed across the study groups (Chi2 = 3.773, P = 0.438). As shown in Table 2 secondary outcome parameters did also not differ between the 3 groups. Figure 1. View largeDownload slide Mean duration (in days) to total oral nutrition (mean ± 1SD) for neonates exposed to rose, vanilla, and placebo. No significant difference was detectable between groups (F = 0.24, P = 0.79). Figure 1. View largeDownload slide Mean duration (in days) to total oral nutrition (mean ± 1SD) for neonates exposed to rose, vanilla, and placebo. No significant difference was detectable between groups (F = 0.24, P = 0.79). Table 2. Primary and secondary outcome parameter (mean ± 1SD) for vanilla and placebo applying a presentation frequency of the “Sniffin’ Sticks” of >66.67% of feeding procedures, Mann–Whitney U test Vanilla n = 18 Placebo n = 13 Rose n = 14 Age corrected by reaching total oral nutrition in days 252 ± 14.3 (P = 0.59) 253 ± 9.1 253 ± 16.3 (P = 0.67) Age by reaching total oral nutrition in days 13 ± 5.0 (P = 0.24) 21 ± 15.6 21 ± 3.1 (P = 0.96) Weight by reaching total oral nutrition in g 2349.4 ± 342.1 (P = 0.79) 2372.4 ± 189.7 2311 ± 432.2 (P = 0.09) Time until total oral nutrition after randomization in days 8 ± 5.4 (P = 0.04) 15 ± 7.3 12 ± 6.6 (P = 0.99) Duration of hospitalization after randomization in days 11 ± 6.5 (P = 0.03) 21 ± 13.7 15 ± 7.3 (P = 0.87) Duration of hospitalization in days 18 ± 6.0 (P = 0.19) 27 ± 16.8 25 ± 13.9 (P = 1.0) Vanilla n = 18 Placebo n = 13 Rose n = 14 Age corrected by reaching total oral nutrition in days 252 ± 14.3 (P = 0.59) 253 ± 9.1 253 ± 16.3 (P = 0.67) Age by reaching total oral nutrition in days 13 ± 5.0 (P = 0.24) 21 ± 15.6 21 ± 3.1 (P = 0.96) Weight by reaching total oral nutrition in g 2349.4 ± 342.1 (P = 0.79) 2372.4 ± 189.7 2311 ± 432.2 (P = 0.09) Time until total oral nutrition after randomization in days 8 ± 5.4 (P = 0.04) 15 ± 7.3 12 ± 6.6 (P = 0.99) Duration of hospitalization after randomization in days 11 ± 6.5 (P = 0.03) 21 ± 13.7 15 ± 7.3 (P = 0.87) Duration of hospitalization in days 18 ± 6.0 (P = 0.19) 27 ± 16.8 25 ± 13.9 (P = 1.0) Significant results are in bold. View Large Table 2. Primary and secondary outcome parameter (mean ± 1SD) for vanilla and placebo applying a presentation frequency of the “Sniffin’ Sticks” of >66.67% of feeding procedures, Mann–Whitney U test Vanilla n = 18 Placebo n = 13 Rose n = 14 Age corrected by reaching total oral nutrition in days 252 ± 14.3 (P = 0.59) 253 ± 9.1 253 ± 16.3 (P = 0.67) Age by reaching total oral nutrition in days 13 ± 5.0 (P = 0.24) 21 ± 15.6 21 ± 3.1 (P = 0.96) Weight by reaching total oral nutrition in g 2349.4 ± 342.1 (P = 0.79) 2372.4 ± 189.7 2311 ± 432.2 (P = 0.09) Time until total oral nutrition after randomization in days 8 ± 5.4 (P = 0.04) 15 ± 7.3 12 ± 6.6 (P = 0.99) Duration of hospitalization after randomization in days 11 ± 6.5 (P = 0.03) 21 ± 13.7 15 ± 7.3 (P = 0.87) Duration of hospitalization in days 18 ± 6.0 (P = 0.19) 27 ± 16.8 25 ± 13.9 (P = 1.0) Vanilla n = 18 Placebo n = 13 Rose n = 14 Age corrected by reaching total oral nutrition in days 252 ± 14.3 (P = 0.59) 253 ± 9.1 253 ± 16.3 (P = 0.67) Age by reaching total oral nutrition in days 13 ± 5.0 (P = 0.24) 21 ± 15.6 21 ± 3.1 (P = 0.96) Weight by reaching total oral nutrition in g 2349.4 ± 342.1 (P = 0.79) 2372.4 ± 189.7 2311 ± 432.2 (P = 0.09) Time until total oral nutrition after randomization in days 8 ± 5.4 (P = 0.04) 15 ± 7.3 12 ± 6.6 (P = 0.99) Duration of hospitalization after randomization in days 11 ± 6.5 (P = 0.03) 21 ± 13.7 15 ± 7.3 (P = 0.87) Duration of hospitalization in days 18 ± 6.0 (P = 0.19) 27 ± 16.8 25 ± 13.9 (P = 1.0) Significant results are in bold. View Large There were however significant group differences in the relation between the “time until total oral nutrition” and the odor presentation frequency. While no effect was observed for the Placebo (r = −0.006, P = 0.969) and the Rose group (r = −0.13, P = 0.398), a significant effect was observed for the Vanilla group. Hence, a higher presentation frequency of vanilla was significantly related to a reduced “time until total oral nutrition” (r = −0.505, P = 0.001). In order to determine the minimal amount of odor presentation for “time until total oral nutrition”, we calculated the presentation time dependent course of effect size. An effect size of d > 0.5 was only observed in cases when the presentation frequency of the vanilla odor was above 60% (Figure 2). Figure 2. View largeDownload slide Effect size of odor presentation (vanilla and rose) in relation to placebo. Figure 2. View largeDownload slide Effect size of odor presentation (vanilla and rose) in relation to placebo. Based on these findings, the analysis was repeated choosing only children with a minimum presentation frequency of the “Sniffin’ Sticks” of two-thirds of the feedings (66.7%). A total of 45 infants met this criterion (vanilla n = 18, rose n = 14, placebo n = 13). In line with the whole study population, baseline parameters did not differ significantly between the study groups in this subpopulation. Furthermore, these infants did not differ significantly in any of the baseline characteristics (weight, length, gestational week, APGAR of 5 min, age, and weight at randomization date). Within this subgroup the distribution of sex, total, or partial gavage feeding at randomization and final feeding method was significant different. Children in the vanilla group had a significantly reduced duration until full oral feeding in comparison to children in the placebo group by 7.2 ± 6.9 days (Mann–Whitney U test: Z = 2.05, P = 0.04) (see Table 2) while no significant difference was found between the study group receiving rose odor and the control group (Mann–Whitney U test: Z = 0.27, P = 0.79). In addition, the duration of hospitalization after randomization was significantly reduced in the vanilla group compared to placebo by 9.2 ± 7.2 days (Mann–Whitney U test: Z = 2.15, P = 0.03). Infants in the vanilla group were discharged after 11.4 ± 6.5 days, while the placebo group left the hospital 20.6 ± 13.7 days after randomization. Children in the rose odor group were discharged from hospital after 14.9 ± 7.3 days, which was not significantly different compared to the control group (Mann–Whitney U test: Z = 0.80, P = 0.43). A logRank test according to Kaplan–Meier was applied showing that children in the vanilla odor group showed a faster transition to total oral feeding compared to children in the control group (P = 0.038) (Figure 3). Figure 3. View largeDownload slide Percentage of infants reaching total oral feeding applying a presentation frequency of the “Sniffin’ Sticks” of >66.67% of feeding procedures. Significant difference between infants exposed to vanilla in comparison to placebo P = 0.038. Figure 3. View largeDownload slide Percentage of infants reaching total oral feeding applying a presentation frequency of the “Sniffin’ Sticks” of >66.67% of feeding procedures. Significant difference between infants exposed to vanilla in comparison to placebo P = 0.038. Discussion In the current study, we showed that odors are of importance in feeding of infants and that odor stimulation with vanilla can lead to a faster transition from gavage to oral feeding. The odorous environment is of importance to the postnatal adaptation and it helps to build a continuum between the prenatal and postnatal surrounding (Schaal et al. 2004). Infants are most likely able to smell from the 28th gestational week in utero and learn about their mother’s nutrition during pregnancy (Chuah and Zheng 1987; Johnson et al. 1995; Mennella et al. 2001). For example, infants whose mothers consumed anise during the final stages of pregnancy preferred the odor of anise more than infants whose mothers avoided the consumption of anise (Schaal 2000). Therefore it is assumed that a fetus is already able to smell components of its mothers diet through the amniotic fluid (Schaal 2000). In the postnatal period, the sense of smell is involved in several behavioral responses of the newborn. It has been shown that the olfactory function is helping the children to find their mothers breast (Varendi et al. 1994). Newborns are already able to discriminate between different odors in their environment (Soussignan et al. 1997), which is of great value to distinguish between an important smell, e.g. the odor of breast milk and an unimportant odor, e.g. rose. Furthermore, breast milk odor is known to increase non-nutritive sucking, which is needed to develop a coordinated suck-swallow-breathing mechanism (Chuah and Zheng 1987; Varendi et al. 1994; Kobal et al. 1996). Due to gavage feeding, infants do not smell the odorants of the breast or formula milk during feeding and therefore the physiological food–odor association is interrupted. The olfactory region of the nose is partially blocked by the nasogastric feeding tube, the milk does not pass through the mouth and pharynx and is administered directly into the stomach. Only through eructating a retronasal odor perception could be possible. As has been shown, the odor of breast milk is a possible method to increase the non-nutritive sucking. It facilitates the weaning from the feeding tube (Yildiz et al. 2011). To use the odor of breast milk to stimulate the oral nutrition turns out to be challenging because of the daily routine on an intensive or intermediate care unit: the milk has to be readily available and the device for milk odor presentation had to be freshly prepared before every feeding. Therefore, it is necessary to use a different food-associated odor, which might have similar effects as breast milk odor. Vanilla is a positively attributed (Soussignan et al. 1997; Bartocci et al. 2000; Jebreili et al. 2015), worldwide used and a food associated odor, which is not known to have any adverse effects (Sinha et al. 2008). Vanilla odor and flavor have already been used in studies with infants and positive effects were reported: reduction of apnea rate (Marlier et al. 2005), a calming effect after painful procedures (Jebreili et al. 2015), increasing of sucking behavior (Mennella and Beauchamp 1996). The other stimulus used in our study was a rose odor, which is thought to be physiologically and psychologically relaxing (Igarashi et al. 2014a, b). Regarding the primary outcome of our study—time from study entry to complete oral feeding—within the whole study population, no significant difference was measureable between the study groups. In addition to secondary outcomes did not lead to significant differences between the study groups. Based on the idea that our odor presentation is resulting in a conditioning of the infants (Schaal et al. 2004), one possible explanation could be that, to some infants, the odor was not presented often enough to connect the perception of an odor with the experience of being fed. In daily routine of an intermediate care unit, a new intervention can be forgotten by the nurses during the care of the infants due to apnea episodes, feeding difficulties, or interruptions through emergencies. The analysis of effect size—by Cohen’s d—of odor stimulation on the lengths of gavage feeding in relation to the odor stimulation frequency showed a steep increase especially for vanilla odor. This finding supports the idea that the odor stimulus must be presented on a high frequency before feeding to be linked to food intake by the infants. Infants who received the vanilla odor presentation on a regular basis showed a significant reduction of the length of tube feeding and hospitalization time after randomization while rose odor administration did not lead to a significant reduction in comparison to placebo. It can be assumed that the reason for the impact of vanilla odor on the feeding of infants is based on a conditioning process, which links the odorant with the procedure of feeding. Previously it was shown that odor-dependent conditioning is possible in early infants (Sullivan et al. 1991). A possible explanation of the higher conditioning influence of vanilla as opposed to rose could be the fact that vanilla is a food associated, highly consumed (Sinha et al. 2008) odor and the likelihood of vanilla consumption of the mothers during pregnancy and postpartum resulted in a familiarity of the infants with vanilla odor (Schaal 2000; Mennella et al. 2001). Even if we did not survey the vanilla intake of the mothers we assume a regular intake by a worldwide consumption of 12,000 t per year (Sinha et al. 2008). In contrast, rose flavor is consumed much less frequently. Another possible explanation is that the odor of vanilla itself is attractive and appetite stimulating for infants. No such effect was reported for rose in any study (Sinha et al. 2008). Adding vanilla flavor to breast milk and formula has already been examined to increase sucking in infants followed by a longer drinking time and higher milk intake (Mennella and Beauchamp 1996). Authors of a recent study reported that the presentation of vanilla odor resulted in a better food tolerance in preterm infants during the first days of enteral nutrition (Beker et al. 2017). This effect could have been also the case in our study leading to increased appetite and better drinking behavior of the infants. Although the results showed a convincing influence of vanilla odor stimulation before feeding on the transition from gavage to oral feeding limitations of the study have to be pointed out. Only a small group of infants met the criterion of an odor presentation frequency above 66.7%. So the remaining sample was much smaller and no longer normally distributed. We paid attention to this fact using non-parametric statistical analysis. Another limitation could be the fact that the study was placebo controlled but not blinded. Both nurses and parents were able to detect from the odor of the “Sniffin’ Stick” into which group the infant was randomized. This could have possibly influenced the expectations and the treatment of the infants. At last we want to mention that no information about the infant’s mothers diet were collected so that we can only assume about the vanilla consumption and fallowing flavor exposure of infants during pregnancy. In conclusion, we were able to observe a benefit of vanilla odor stimulation before feeding not only on the transition from gavage to oral feeding but also on the length of hospitalization. Based on this, larger studies need to be conducted with the idea to investigate this procedure for its usefulness in clinical routine. In our current investigation, we were not able to study the mechanisms underlying the positive effect of vanilla odor stimulation. This will be the target of future research. Funding This study was supported by a grant from the Else Kröner-Fresenius-Stiftung to VAS and from the Deutsche Forschungsgemeinschaft to TH (DFG HU441-18-1). Conflict of interest None References Bartocci M , Winberg J , Ruggiero C , Bergqvist LL , Serra G , Lagercrantz H . 2000 . Activation of olfactory cortex in newborn infants after odor stimulation: a functional near-infrared spectroscopy study . Pediatr Res . 48 : 18 – 23 . Google Scholar CrossRef Search ADS PubMed Beker F , Opie G , Noble E , Jiang Y , Bloomfield FH . 2017 . Smell and taste to improve nutrition in very preterm infants: a randomized controlled pilot trial . Neonatology . 111 : 260 – 266 . Google Scholar CrossRef Search ADS PubMed Bingham PM , Abassi S , Sivieri E . 2003 . A pilot study of milk odor effect on nonnutritive sucking by premature newborns . Arch Pediatr Adolesc Med . 157 : 72 – 75 . Google Scholar CrossRef Search ADS PubMed Bingham PM , Churchill D , Ashikaga T . 2007 . Breast milk odor via olfactometer for tube-fed, premature infants . Behav Res Methods . 39 : 630 – 634 . Google Scholar CrossRef Search ADS PubMed Calkins KL , Venick RS , Devaskar SU . 2014 . Complications associated with parenteral nutrition in the neonate . Clin Perinatol . 41 : 331 – 345 . Google Scholar CrossRef Search ADS PubMed Chuah MI , Zheng DR . 1987 . Olfactory marker protein is present in olfactory receptor cells of human fetuses . Neuroscience . 23 : 363 – 370 . Google Scholar CrossRef Search ADS PubMed Delaney AL , Arvedson JC . 2008 . Development of swallowing and feeding: prenatal through first year of life . Dev Disabil Res Rev . 14 : 105 – 117 . Google Scholar CrossRef Search ADS PubMed Gewolb IH , Vice FL . 2006 . Maturational changes in the rhythms, patterning, and coordination of respiration and swallow during feeding in preterm and term infants . Dev Med Child Neurol . 48 : 589 – 594 . Google Scholar CrossRef Search ADS PubMed Igarashi M , Ikei H , Song C , Miyazaki Y . 2014a . Effects of olfactory stimulation with rose and orange oil on prefrontal cortex activity . Complement Ther Med . 22 : 1027 – 1031 . Google Scholar CrossRef Search ADS Igarashi M , Song C , Ikei H , Ohira T , Miyazaki Y . 2014b . Effect of olfactory stimulation by fresh rose flowers on autonomic nervous activity . J Altern Complement Med . 20 : 727 – 731 . Google Scholar CrossRef Search ADS Jebreili M , Neshat H , Seyyedrasouli A , Ghojazade M , Hosseini MB , Hamishehkar H . 2015 . Comparison of breastmilk odor and vanilla odor on mitigating premature infants’ response to pain during and after venipuncture . Breastfeed Med . 10 : 362 – 365 . Google Scholar CrossRef Search ADS PubMed Johnson EW , Eller PM , Jafek BW . 1995 . Distribution of OMP-, PGP 9.5- and CaBP-like immunoreactive chemoreceptor neurons in the developing human olfactory epithelium . Anat Embryol (Berl) . 191 : 311 – 317 . Google Scholar CrossRef Search ADS PubMed Kobal G , Hummel T , Sekinger B , Barz S , Roscher S , Wolf S . 1996 . “Sniffin’ sticks”: screening of olfactory performance . Rhinology . 34 : 222 – 226 . Google Scholar PubMed Marlier L , Gaugler C , Messer J . 2005 . Olfactory stimulation prevents apnea in premature newborns . Pediatrics . 115 : 83 – 88 . Google Scholar CrossRef Search ADS PubMed Mennella JA , Beauchamp GK . 1996 . The human infants’ response to vanilla flavors in mother’s milk and formula . Infant Behav Dev . 19 : 13 – 19 . Google Scholar CrossRef Search ADS Mennella JA , Jagnow CP , Beauchamp GK . 2001 . Prenatal and postnatal flavor learning by human infants . Pediatrics . 107 : E88 . Google Scholar CrossRef Search ADS PubMed Russell MJ . 1976 . Human olfactory communication . Nature . 260 : 520 – 522 . Google Scholar CrossRef Search ADS PubMed Schaal B , Hummel T , Soussignan R . 2004 . Olfaction in the fetal and premature infant: functional status and clinical implications . Clin Perinatol . 31 : 261 – 85 , vi. Google Scholar CrossRef Search ADS PubMed Schaal B , Marlier L , Soussignan R . 2000 . Human foetuses learn odours from their pregnant mother’s diet . Chem Senses . 25 : 729 – 737 . Google Scholar CrossRef Search ADS PubMed Sinha AK , Sharma UK , Sharma N . 2008 . A comprehensive review on vanilla flavor: extraction, isolation and quantification of vanillin and others constituents . Int J Food Sci Nutr . 59 : 299 – 326 . doi:10.1080/09687630701539350 Google Scholar CrossRef Search ADS PubMed Soussignan R , Schaal B , Marlier L , Jiang T . 1997 . Facial and autonomic responses to biological and artificial olfactory stimuli in human neonates: re-examining early hedonic discrimination of odors . Physiol Behav . 62 : 745 – 758 . Google Scholar CrossRef Search ADS PubMed Sullivan RM , Taborsky-Barba S , Mendoza R , Itano A , Leon M , Cotman CW , Payne TF , Lott I . 1991 . Olfactory classical conditioning in neonates . Pediatrics . 87 : 511 – 518 . Google Scholar PubMed Varendi H , Porter RH , Winberg J . 1994 . Does the newborn baby find the nipple by smell ? Lancet . 344 : 989 – 990 . Google Scholar CrossRef Search ADS PubMed Yildiz, A, Arikan, D, Gözüm, S, Taştekın, A, Budancamanak, İ. 2011. The effect of the odor of breast milk on the time needed for transition from gavage to total oral feeding in preterm infants: the odor of breast milk. J Nurs Scholarsh. 43:265–273. doi:10.1111/j.1547-5069.2011.01410.x © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Chemical Senses Oxford University Press

Sniffin’ Away the Feeding Tube: The Influence of Olfactory Stimulation on Oral Food Intake in Newborns and Premature Infants

Chemical Senses , Volume 43 (7) – Sep 1, 2018

Loading next page...
 
/lp/ou_press/sniffin-away-the-feeding-tube-the-influence-of-olfactory-stimulation-RJaQv5QFXn
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0379-864X
eISSN
1464-3553
D.O.I.
10.1093/chemse/bjy034
Publisher site
See Article on Publisher Site

Abstract

Abstract Because of their immaturity, many premature infants are fed via nasogastric tube. One objective of the neonatal care is to feed infants orally early. The olfactory function of premature infants is developed before birth and odorants have a significant impact on nutrition in infants. The aim of the study was to test whether odor stimulation has a positive effect on the transition from gavage to oral feeding in infants. Participants were premature infants with gestational age of more than 27 weeks, with full or partial gavage feeding, stable vital parameters and without invasive ventilation. Before each feeding procedure an odorant was presented in front of the infant’s nose. Infants were randomized into 1 of 3 groups and received either rose odor (not food-associated), vanilla odor (food-associated), or placebo (no odor). The primary outcome of the study was defined as the time until complete oral nutrition. About 150 children born at a postnatal age of 9.5 ± 7.8 days were included in this study. The duration until complete oral nutrition was reached after 11.8 ± 7.7 (vanilla), 12.2 ± 7.7 (rose), and 12.9 ± 8.8 (control) days. A nearly linear relation between odor presentation frequency and effect size was detectable. For infants that received the intervention for more than 66.7% of the time the length of gavage feeding (8 ± 5.4) and hospitalization (11 ± 6.5) was significantly lower in the vanilla group when compared with control (15 ± 7.3 and 21 ± 13.7, respectively). Odor stimulation with vanilla has an impact on oral feeding in premature infants, however the odor has to be presented on regular basis. gavage, nutrition, olfaction, premature infant Introduction The clinical and scientific progress in neonatology continuously improves the survival of premature infants especially with very low birth weight. Therefore, the focus is shifting to improve therapy and outcome. One major target is an adequate and early enteral feeding of premature infants. The majority of premature infants need to be fed by gavage due to insufficient sucking and a lack of suck–swallow–breath coordination (Gewolb and Vice 2006). As a result of gavage-feeding infants are not able to experience the flavor, a combination of smell and taste, of breast or formula milk. The physiological connection between the odor of nutriment, which initiates the cephalic phase, followed by sucking behavior and the satisfactory feeling of gastric filling is interrupted. Although gavage feeding is necessary and superior to parenteral nutrition (Calkins et al. 2014), the lack of olfactory perception during feeding might result in a prolonged transition from gavage to oral feeding. The importance of olfaction for newborns, especially of mother-associated odorants is well-known (Russell 1976; Varendi et al. 1994). Olfactory receptor neurons were found in preterm infants as early as 24–27 weeks of gestation (Chuah and Zheng 1987; Johnson et al. 1995). Therefore, it can be assumed that the olfactory sense is functioning in preterm infants. In 1994, a study showed the preference of infants for the odor of their mothers’ breast through head-movement towards the unwashed mother’s nipple (Varendi et al. 1994). Food-associated odorants are also important for the first days after birth. For example, 2 studies demonstrated an increase of non-nutritive sucking in response to milk odorant (Bingham et al. 2003, 2007). Non-nutritive-sucking is known to be important for infants to learn a coordinated suck–swallow–breathing procedure (Delaney and Arvedson 2008). Finally, a pilot study found a positive effect of breast milk odor on the duration of gavage feeding and hospitalization time of preterm infants born between gestation weeks 28–34 (Yildiz et al. 2011). Caused by the poor practicability of presenting fresh breastmilk odor in daily routine in an intermediate care unit, it was necessary to select another food-associated odor for this study. Stimulation with vanilla odor is not only food-associated, side effect-free and safe but has already been used in studies with infants showing an increase of sucking behavior (Mennella and Beauchamp 1996) and other positive effects, e.g. a reduction of apnea (Marlier et al. 2005). Based on previous literature, the question arises whether olfactory stimulation in general or only food associated odors support transition from gavage to full oral feeding in premature infants. Therefore, the study was designed using rose and vanilla odor to evaluate whether an intervention of odor stimulation before feeding compared to a control group leads to faster transition from gavage to complete oral feeding. Material and methods This prospective randomized controlled study was approved by the local Ethics Committee of the Medical Faculty at the TU Dresden (EK 356092014). The study was conducted in accordance with the Declaration of Helsinki on Biomedical Studies Involving Human Subjects. Participants Participants were premature infants born at a gestational age of more than 27 weeks with stable vital parameters and without invasive ventilation or CPAP at time of randomization. All newborns received complete or partial gavage feeding. Exclusion criteria were neurological disabilities, intracerebral hemorrhage >II°, confirmed diagnosis of a syndromal disease or anatomical gastrointestinal anomalies. They were recruited shortly after transition from intensive care to intermediate care unit. Written informed consent was obtained from the parents or legal guardians of all participating infants before inclusion in the study. Intervention Children were randomized into 1 of the following 3 study groups: (i) food-related odor “vanilla”, (ii) non-food associated odor “rose”, or (iii) an odorless probe (control). Using a predefined study list, controlling for gestational age children were randomized. Odors were presented to the infants before each feeding. The “Sniffin’ Sticks” were used for odor presentation (Kobal et al. 1996). For the odor presentation, the cap of the “Sniffin’ Sticks” was removed by the nurse and the tip of the pen was positioned for ~10 s ~2 cm under the nostrils of the infants. This procedure was repeated before each feeding by bottle or by nasogastric tube but not before breastfeeding in order to avoid any interference with the physiological mother–child interaction. After each odor presentation, the nurse documented the procedure in an electronic patient data management system (Integrated Care Manager, Dräger AG&co). Each infant received its own “Sniffin’ Stick” (Kobal et al. 1996). “Sniffin’ Sticks” are felt-tip pens, which are filled with an odorant instead of pigment (“Sniffin’ Sticks”, Burghart GmbH). The following substances were used as odors: Vanilla (4-Hdroxy-3-Methoxybenzaldehyd, Sigma&Aldrich #V1104) in solution with 1,2-Propanediol (Sigma&Aldrich, #134368; 3 g in 100 mL), Rose (Frey&Lau, #P0604034, diluted 1:10 in 1,2-Propanediol), the control group (placebo) received an odorless pen. Outcome The primary outcome of the study was defined as duration between study entry and complete oral food intake defined as solely oral feeding for at least 24 h. Body weight, amount of nutrition by gavage-, bottle- and breastfeeding was recorded daily until the end of hospitalization. Furthermore, a prospectively defined subanalysis was performed to determine the effect of percentage of odor presentation. Statistical analysis Analyses were performed using IBM Statistical Package for the Social Sciences version 23.0 (SPSS Inc.) software with significance set as P < 0.05. One-way ANOVA was used to test differences between the 3 study groups. Between-group comparisons were performed using Bonferroni-corrected posthoc tests. Whenever appropriate non-parametric tests were used for comparing not normally distributed samples. In order to determine the required minimal presentation time, the presentation time dependent course of effect size was calculated. This was done by the following routine: those 16 participants who received the odor stimulation for most time per odor were first averaged and the effect size was determined between baseline and outcome measurement. Sixteen participants were chosen in order to ensure a robust sample size for effect size calculation. For rose odor, this compared the presentation time from 63 to 100%; for vanilla odor this compared the time between 71 and 100 %. In the next step, we consecutively calculated the effect size for the next lower odor presentation time, hence 17 participants were included into the analysis. Thereafter, we included 128 participants and so on until the final effect size calculation comprised the whole sample. The minimal presentation time was plotted against the effect size per odor. A statistical calculation determining the necessary sample size was done a priory. For the comparison of the effect of odor stimulation on the transition from gavage to oral feeding between 3 groups an ANOVA should be applied. To obtain valid results with an effect size of f = 0.25 at an alpha level of 0.05 and a power of ß = 0.75 a study population of 141 infants would be necessary. Therefore, we decided to include 50 infants in each of the 3 study groups. Results A total of 150 healthy infants participated in the study, 15 children had to be excluded from the analysis for the following reasons: withdrawal (n = 3), transferal to other hospitals (n = 12). No infant had to be excluded due to adverse effects of the odor stimulation. The details of the baseline and birth data of the infants are presented in detail in Table 1. Table 1. Baseline characteristics of the 3 study groups: rose, vanilla, and placebo, n amount of children in the studygroup, mean ± SD Rose Vanilla Placebo Total Weight at birth in g 1862.5 ± 448.63 n = 46 1866.5 ± 424.6 n = 49 1889.9 ± 468.4 n = 40 1872.1 ± 442.9 n = 135 Length at birth in cm 43 ± 3.29 n = 46 43.3 ± 2.99 n = 49 42.9 ± 3.58 n = 40 43.1 ± 3.26 n = 135 Gestational week in days 231 ± 16.9 n = 46 230 ± 16.2 n = 49 230 ± 16.1 n = 40 230 ± 16.3 n = 135 APGAR 5 min 7.8 ± 1.12 n = 46 8.3 ± 0.93 n = 48 8.2 ± 1.15 n = 38 8.1 ± 1.08 n = 132 Age at randomization in days 9.8 ± 8.66 n = 46 9.2 ± 6.29 n = 49 9.4 ± 8.62 n = 40 9.5 ± 7.81 n = 135 Weight at randomization in g 1956.6 ± 411.76 n = 46 1916.5 ± 448.09 n = 48 1953.5 ± 384.98 n = 40 1941.3 ± 414.88 n = 134 total of gavage fed infants at randomization 2 2 2 6 % oral feeding at randomization 25.0 ± 20.78 n = 46 25.9 ± 21.0 n = 49 24.7 ± 18.69 n = 39 25.3 ± 20.13 n = 134 Rose Vanilla Placebo Total Weight at birth in g 1862.5 ± 448.63 n = 46 1866.5 ± 424.6 n = 49 1889.9 ± 468.4 n = 40 1872.1 ± 442.9 n = 135 Length at birth in cm 43 ± 3.29 n = 46 43.3 ± 2.99 n = 49 42.9 ± 3.58 n = 40 43.1 ± 3.26 n = 135 Gestational week in days 231 ± 16.9 n = 46 230 ± 16.2 n = 49 230 ± 16.1 n = 40 230 ± 16.3 n = 135 APGAR 5 min 7.8 ± 1.12 n = 46 8.3 ± 0.93 n = 48 8.2 ± 1.15 n = 38 8.1 ± 1.08 n = 132 Age at randomization in days 9.8 ± 8.66 n = 46 9.2 ± 6.29 n = 49 9.4 ± 8.62 n = 40 9.5 ± 7.81 n = 135 Weight at randomization in g 1956.6 ± 411.76 n = 46 1916.5 ± 448.09 n = 48 1953.5 ± 384.98 n = 40 1941.3 ± 414.88 n = 134 total of gavage fed infants at randomization 2 2 2 6 % oral feeding at randomization 25.0 ± 20.78 n = 46 25.9 ± 21.0 n = 49 24.7 ± 18.69 n = 39 25.3 ± 20.13 n = 134 View Large Table 1. Baseline characteristics of the 3 study groups: rose, vanilla, and placebo, n amount of children in the studygroup, mean ± SD Rose Vanilla Placebo Total Weight at birth in g 1862.5 ± 448.63 n = 46 1866.5 ± 424.6 n = 49 1889.9 ± 468.4 n = 40 1872.1 ± 442.9 n = 135 Length at birth in cm 43 ± 3.29 n = 46 43.3 ± 2.99 n = 49 42.9 ± 3.58 n = 40 43.1 ± 3.26 n = 135 Gestational week in days 231 ± 16.9 n = 46 230 ± 16.2 n = 49 230 ± 16.1 n = 40 230 ± 16.3 n = 135 APGAR 5 min 7.8 ± 1.12 n = 46 8.3 ± 0.93 n = 48 8.2 ± 1.15 n = 38 8.1 ± 1.08 n = 132 Age at randomization in days 9.8 ± 8.66 n = 46 9.2 ± 6.29 n = 49 9.4 ± 8.62 n = 40 9.5 ± 7.81 n = 135 Weight at randomization in g 1956.6 ± 411.76 n = 46 1916.5 ± 448.09 n = 48 1953.5 ± 384.98 n = 40 1941.3 ± 414.88 n = 134 total of gavage fed infants at randomization 2 2 2 6 % oral feeding at randomization 25.0 ± 20.78 n = 46 25.9 ± 21.0 n = 49 24.7 ± 18.69 n = 39 25.3 ± 20.13 n = 134 Rose Vanilla Placebo Total Weight at birth in g 1862.5 ± 448.63 n = 46 1866.5 ± 424.6 n = 49 1889.9 ± 468.4 n = 40 1872.1 ± 442.9 n = 135 Length at birth in cm 43 ± 3.29 n = 46 43.3 ± 2.99 n = 49 42.9 ± 3.58 n = 40 43.1 ± 3.26 n = 135 Gestational week in days 231 ± 16.9 n = 46 230 ± 16.2 n = 49 230 ± 16.1 n = 40 230 ± 16.3 n = 135 APGAR 5 min 7.8 ± 1.12 n = 46 8.3 ± 0.93 n = 48 8.2 ± 1.15 n = 38 8.1 ± 1.08 n = 132 Age at randomization in days 9.8 ± 8.66 n = 46 9.2 ± 6.29 n = 49 9.4 ± 8.62 n = 40 9.5 ± 7.81 n = 135 Weight at randomization in g 1956.6 ± 411.76 n = 46 1916.5 ± 448.09 n = 48 1953.5 ± 384.98 n = 40 1941.3 ± 414.88 n = 134 total of gavage fed infants at randomization 2 2 2 6 % oral feeding at randomization 25.0 ± 20.78 n = 46 25.9 ± 21.0 n = 49 24.7 ± 18.69 n = 39 25.3 ± 20.13 n = 134 View Large First, we tested whether there was a significant effect of group of presentation and whether there was an interaction between odor presentation time and group. We therefore conducted a univariate ANOVA (depend variable “time until total oral nutrition after randomization in days”) with the fixed effect of group and the covariate of odor presentation frequency and modeled the main effect of group and the interaction group by time until total oral nutrition after randomization in days. In results, there was a trend for group (F(2,120) = 2.469, P = 0.089) and a significant interaction effect (F(1,120) = 3.921, P = 0.01). In order to explore the interaction further, the results were analyzed more in detail. At baseline, the 3 study groups did not show a significant difference (Table 1). The duration between study entry and full oral feeding was 11.8 ± 7.7, 12.2 ± 7.7, and 12.8 ± 8.8 days for the vanilla, rose, and control group, respectively (Figure 1). Thus, the time of presentation was not significantly different between the 3 groups (F(2,123) = 0.24, P = 0.79). Taking a look at the feeding method (breast milk, formula, partial formula) at discharge from the hospital the feeding methods were equally distributed across the study groups (Chi2 = 3.773, P = 0.438). As shown in Table 2 secondary outcome parameters did also not differ between the 3 groups. Figure 1. View largeDownload slide Mean duration (in days) to total oral nutrition (mean ± 1SD) for neonates exposed to rose, vanilla, and placebo. No significant difference was detectable between groups (F = 0.24, P = 0.79). Figure 1. View largeDownload slide Mean duration (in days) to total oral nutrition (mean ± 1SD) for neonates exposed to rose, vanilla, and placebo. No significant difference was detectable between groups (F = 0.24, P = 0.79). Table 2. Primary and secondary outcome parameter (mean ± 1SD) for vanilla and placebo applying a presentation frequency of the “Sniffin’ Sticks” of >66.67% of feeding procedures, Mann–Whitney U test Vanilla n = 18 Placebo n = 13 Rose n = 14 Age corrected by reaching total oral nutrition in days 252 ± 14.3 (P = 0.59) 253 ± 9.1 253 ± 16.3 (P = 0.67) Age by reaching total oral nutrition in days 13 ± 5.0 (P = 0.24) 21 ± 15.6 21 ± 3.1 (P = 0.96) Weight by reaching total oral nutrition in g 2349.4 ± 342.1 (P = 0.79) 2372.4 ± 189.7 2311 ± 432.2 (P = 0.09) Time until total oral nutrition after randomization in days 8 ± 5.4 (P = 0.04) 15 ± 7.3 12 ± 6.6 (P = 0.99) Duration of hospitalization after randomization in days 11 ± 6.5 (P = 0.03) 21 ± 13.7 15 ± 7.3 (P = 0.87) Duration of hospitalization in days 18 ± 6.0 (P = 0.19) 27 ± 16.8 25 ± 13.9 (P = 1.0) Vanilla n = 18 Placebo n = 13 Rose n = 14 Age corrected by reaching total oral nutrition in days 252 ± 14.3 (P = 0.59) 253 ± 9.1 253 ± 16.3 (P = 0.67) Age by reaching total oral nutrition in days 13 ± 5.0 (P = 0.24) 21 ± 15.6 21 ± 3.1 (P = 0.96) Weight by reaching total oral nutrition in g 2349.4 ± 342.1 (P = 0.79) 2372.4 ± 189.7 2311 ± 432.2 (P = 0.09) Time until total oral nutrition after randomization in days 8 ± 5.4 (P = 0.04) 15 ± 7.3 12 ± 6.6 (P = 0.99) Duration of hospitalization after randomization in days 11 ± 6.5 (P = 0.03) 21 ± 13.7 15 ± 7.3 (P = 0.87) Duration of hospitalization in days 18 ± 6.0 (P = 0.19) 27 ± 16.8 25 ± 13.9 (P = 1.0) Significant results are in bold. View Large Table 2. Primary and secondary outcome parameter (mean ± 1SD) for vanilla and placebo applying a presentation frequency of the “Sniffin’ Sticks” of >66.67% of feeding procedures, Mann–Whitney U test Vanilla n = 18 Placebo n = 13 Rose n = 14 Age corrected by reaching total oral nutrition in days 252 ± 14.3 (P = 0.59) 253 ± 9.1 253 ± 16.3 (P = 0.67) Age by reaching total oral nutrition in days 13 ± 5.0 (P = 0.24) 21 ± 15.6 21 ± 3.1 (P = 0.96) Weight by reaching total oral nutrition in g 2349.4 ± 342.1 (P = 0.79) 2372.4 ± 189.7 2311 ± 432.2 (P = 0.09) Time until total oral nutrition after randomization in days 8 ± 5.4 (P = 0.04) 15 ± 7.3 12 ± 6.6 (P = 0.99) Duration of hospitalization after randomization in days 11 ± 6.5 (P = 0.03) 21 ± 13.7 15 ± 7.3 (P = 0.87) Duration of hospitalization in days 18 ± 6.0 (P = 0.19) 27 ± 16.8 25 ± 13.9 (P = 1.0) Vanilla n = 18 Placebo n = 13 Rose n = 14 Age corrected by reaching total oral nutrition in days 252 ± 14.3 (P = 0.59) 253 ± 9.1 253 ± 16.3 (P = 0.67) Age by reaching total oral nutrition in days 13 ± 5.0 (P = 0.24) 21 ± 15.6 21 ± 3.1 (P = 0.96) Weight by reaching total oral nutrition in g 2349.4 ± 342.1 (P = 0.79) 2372.4 ± 189.7 2311 ± 432.2 (P = 0.09) Time until total oral nutrition after randomization in days 8 ± 5.4 (P = 0.04) 15 ± 7.3 12 ± 6.6 (P = 0.99) Duration of hospitalization after randomization in days 11 ± 6.5 (P = 0.03) 21 ± 13.7 15 ± 7.3 (P = 0.87) Duration of hospitalization in days 18 ± 6.0 (P = 0.19) 27 ± 16.8 25 ± 13.9 (P = 1.0) Significant results are in bold. View Large There were however significant group differences in the relation between the “time until total oral nutrition” and the odor presentation frequency. While no effect was observed for the Placebo (r = −0.006, P = 0.969) and the Rose group (r = −0.13, P = 0.398), a significant effect was observed for the Vanilla group. Hence, a higher presentation frequency of vanilla was significantly related to a reduced “time until total oral nutrition” (r = −0.505, P = 0.001). In order to determine the minimal amount of odor presentation for “time until total oral nutrition”, we calculated the presentation time dependent course of effect size. An effect size of d > 0.5 was only observed in cases when the presentation frequency of the vanilla odor was above 60% (Figure 2). Figure 2. View largeDownload slide Effect size of odor presentation (vanilla and rose) in relation to placebo. Figure 2. View largeDownload slide Effect size of odor presentation (vanilla and rose) in relation to placebo. Based on these findings, the analysis was repeated choosing only children with a minimum presentation frequency of the “Sniffin’ Sticks” of two-thirds of the feedings (66.7%). A total of 45 infants met this criterion (vanilla n = 18, rose n = 14, placebo n = 13). In line with the whole study population, baseline parameters did not differ significantly between the study groups in this subpopulation. Furthermore, these infants did not differ significantly in any of the baseline characteristics (weight, length, gestational week, APGAR of 5 min, age, and weight at randomization date). Within this subgroup the distribution of sex, total, or partial gavage feeding at randomization and final feeding method was significant different. Children in the vanilla group had a significantly reduced duration until full oral feeding in comparison to children in the placebo group by 7.2 ± 6.9 days (Mann–Whitney U test: Z = 2.05, P = 0.04) (see Table 2) while no significant difference was found between the study group receiving rose odor and the control group (Mann–Whitney U test: Z = 0.27, P = 0.79). In addition, the duration of hospitalization after randomization was significantly reduced in the vanilla group compared to placebo by 9.2 ± 7.2 days (Mann–Whitney U test: Z = 2.15, P = 0.03). Infants in the vanilla group were discharged after 11.4 ± 6.5 days, while the placebo group left the hospital 20.6 ± 13.7 days after randomization. Children in the rose odor group were discharged from hospital after 14.9 ± 7.3 days, which was not significantly different compared to the control group (Mann–Whitney U test: Z = 0.80, P = 0.43). A logRank test according to Kaplan–Meier was applied showing that children in the vanilla odor group showed a faster transition to total oral feeding compared to children in the control group (P = 0.038) (Figure 3). Figure 3. View largeDownload slide Percentage of infants reaching total oral feeding applying a presentation frequency of the “Sniffin’ Sticks” of >66.67% of feeding procedures. Significant difference between infants exposed to vanilla in comparison to placebo P = 0.038. Figure 3. View largeDownload slide Percentage of infants reaching total oral feeding applying a presentation frequency of the “Sniffin’ Sticks” of >66.67% of feeding procedures. Significant difference between infants exposed to vanilla in comparison to placebo P = 0.038. Discussion In the current study, we showed that odors are of importance in feeding of infants and that odor stimulation with vanilla can lead to a faster transition from gavage to oral feeding. The odorous environment is of importance to the postnatal adaptation and it helps to build a continuum between the prenatal and postnatal surrounding (Schaal et al. 2004). Infants are most likely able to smell from the 28th gestational week in utero and learn about their mother’s nutrition during pregnancy (Chuah and Zheng 1987; Johnson et al. 1995; Mennella et al. 2001). For example, infants whose mothers consumed anise during the final stages of pregnancy preferred the odor of anise more than infants whose mothers avoided the consumption of anise (Schaal 2000). Therefore it is assumed that a fetus is already able to smell components of its mothers diet through the amniotic fluid (Schaal 2000). In the postnatal period, the sense of smell is involved in several behavioral responses of the newborn. It has been shown that the olfactory function is helping the children to find their mothers breast (Varendi et al. 1994). Newborns are already able to discriminate between different odors in their environment (Soussignan et al. 1997), which is of great value to distinguish between an important smell, e.g. the odor of breast milk and an unimportant odor, e.g. rose. Furthermore, breast milk odor is known to increase non-nutritive sucking, which is needed to develop a coordinated suck-swallow-breathing mechanism (Chuah and Zheng 1987; Varendi et al. 1994; Kobal et al. 1996). Due to gavage feeding, infants do not smell the odorants of the breast or formula milk during feeding and therefore the physiological food–odor association is interrupted. The olfactory region of the nose is partially blocked by the nasogastric feeding tube, the milk does not pass through the mouth and pharynx and is administered directly into the stomach. Only through eructating a retronasal odor perception could be possible. As has been shown, the odor of breast milk is a possible method to increase the non-nutritive sucking. It facilitates the weaning from the feeding tube (Yildiz et al. 2011). To use the odor of breast milk to stimulate the oral nutrition turns out to be challenging because of the daily routine on an intensive or intermediate care unit: the milk has to be readily available and the device for milk odor presentation had to be freshly prepared before every feeding. Therefore, it is necessary to use a different food-associated odor, which might have similar effects as breast milk odor. Vanilla is a positively attributed (Soussignan et al. 1997; Bartocci et al. 2000; Jebreili et al. 2015), worldwide used and a food associated odor, which is not known to have any adverse effects (Sinha et al. 2008). Vanilla odor and flavor have already been used in studies with infants and positive effects were reported: reduction of apnea rate (Marlier et al. 2005), a calming effect after painful procedures (Jebreili et al. 2015), increasing of sucking behavior (Mennella and Beauchamp 1996). The other stimulus used in our study was a rose odor, which is thought to be physiologically and psychologically relaxing (Igarashi et al. 2014a, b). Regarding the primary outcome of our study—time from study entry to complete oral feeding—within the whole study population, no significant difference was measureable between the study groups. In addition to secondary outcomes did not lead to significant differences between the study groups. Based on the idea that our odor presentation is resulting in a conditioning of the infants (Schaal et al. 2004), one possible explanation could be that, to some infants, the odor was not presented often enough to connect the perception of an odor with the experience of being fed. In daily routine of an intermediate care unit, a new intervention can be forgotten by the nurses during the care of the infants due to apnea episodes, feeding difficulties, or interruptions through emergencies. The analysis of effect size—by Cohen’s d—of odor stimulation on the lengths of gavage feeding in relation to the odor stimulation frequency showed a steep increase especially for vanilla odor. This finding supports the idea that the odor stimulus must be presented on a high frequency before feeding to be linked to food intake by the infants. Infants who received the vanilla odor presentation on a regular basis showed a significant reduction of the length of tube feeding and hospitalization time after randomization while rose odor administration did not lead to a significant reduction in comparison to placebo. It can be assumed that the reason for the impact of vanilla odor on the feeding of infants is based on a conditioning process, which links the odorant with the procedure of feeding. Previously it was shown that odor-dependent conditioning is possible in early infants (Sullivan et al. 1991). A possible explanation of the higher conditioning influence of vanilla as opposed to rose could be the fact that vanilla is a food associated, highly consumed (Sinha et al. 2008) odor and the likelihood of vanilla consumption of the mothers during pregnancy and postpartum resulted in a familiarity of the infants with vanilla odor (Schaal 2000; Mennella et al. 2001). Even if we did not survey the vanilla intake of the mothers we assume a regular intake by a worldwide consumption of 12,000 t per year (Sinha et al. 2008). In contrast, rose flavor is consumed much less frequently. Another possible explanation is that the odor of vanilla itself is attractive and appetite stimulating for infants. No such effect was reported for rose in any study (Sinha et al. 2008). Adding vanilla flavor to breast milk and formula has already been examined to increase sucking in infants followed by a longer drinking time and higher milk intake (Mennella and Beauchamp 1996). Authors of a recent study reported that the presentation of vanilla odor resulted in a better food tolerance in preterm infants during the first days of enteral nutrition (Beker et al. 2017). This effect could have been also the case in our study leading to increased appetite and better drinking behavior of the infants. Although the results showed a convincing influence of vanilla odor stimulation before feeding on the transition from gavage to oral feeding limitations of the study have to be pointed out. Only a small group of infants met the criterion of an odor presentation frequency above 66.7%. So the remaining sample was much smaller and no longer normally distributed. We paid attention to this fact using non-parametric statistical analysis. Another limitation could be the fact that the study was placebo controlled but not blinded. Both nurses and parents were able to detect from the odor of the “Sniffin’ Stick” into which group the infant was randomized. This could have possibly influenced the expectations and the treatment of the infants. At last we want to mention that no information about the infant’s mothers diet were collected so that we can only assume about the vanilla consumption and fallowing flavor exposure of infants during pregnancy. In conclusion, we were able to observe a benefit of vanilla odor stimulation before feeding not only on the transition from gavage to oral feeding but also on the length of hospitalization. Based on this, larger studies need to be conducted with the idea to investigate this procedure for its usefulness in clinical routine. In our current investigation, we were not able to study the mechanisms underlying the positive effect of vanilla odor stimulation. This will be the target of future research. Funding This study was supported by a grant from the Else Kröner-Fresenius-Stiftung to VAS and from the Deutsche Forschungsgemeinschaft to TH (DFG HU441-18-1). Conflict of interest None References Bartocci M , Winberg J , Ruggiero C , Bergqvist LL , Serra G , Lagercrantz H . 2000 . Activation of olfactory cortex in newborn infants after odor stimulation: a functional near-infrared spectroscopy study . Pediatr Res . 48 : 18 – 23 . Google Scholar CrossRef Search ADS PubMed Beker F , Opie G , Noble E , Jiang Y , Bloomfield FH . 2017 . Smell and taste to improve nutrition in very preterm infants: a randomized controlled pilot trial . Neonatology . 111 : 260 – 266 . Google Scholar CrossRef Search ADS PubMed Bingham PM , Abassi S , Sivieri E . 2003 . A pilot study of milk odor effect on nonnutritive sucking by premature newborns . Arch Pediatr Adolesc Med . 157 : 72 – 75 . Google Scholar CrossRef Search ADS PubMed Bingham PM , Churchill D , Ashikaga T . 2007 . Breast milk odor via olfactometer for tube-fed, premature infants . Behav Res Methods . 39 : 630 – 634 . Google Scholar CrossRef Search ADS PubMed Calkins KL , Venick RS , Devaskar SU . 2014 . Complications associated with parenteral nutrition in the neonate . Clin Perinatol . 41 : 331 – 345 . Google Scholar CrossRef Search ADS PubMed Chuah MI , Zheng DR . 1987 . Olfactory marker protein is present in olfactory receptor cells of human fetuses . Neuroscience . 23 : 363 – 370 . Google Scholar CrossRef Search ADS PubMed Delaney AL , Arvedson JC . 2008 . Development of swallowing and feeding: prenatal through first year of life . Dev Disabil Res Rev . 14 : 105 – 117 . Google Scholar CrossRef Search ADS PubMed Gewolb IH , Vice FL . 2006 . Maturational changes in the rhythms, patterning, and coordination of respiration and swallow during feeding in preterm and term infants . Dev Med Child Neurol . 48 : 589 – 594 . Google Scholar CrossRef Search ADS PubMed Igarashi M , Ikei H , Song C , Miyazaki Y . 2014a . Effects of olfactory stimulation with rose and orange oil on prefrontal cortex activity . Complement Ther Med . 22 : 1027 – 1031 . Google Scholar CrossRef Search ADS Igarashi M , Song C , Ikei H , Ohira T , Miyazaki Y . 2014b . Effect of olfactory stimulation by fresh rose flowers on autonomic nervous activity . J Altern Complement Med . 20 : 727 – 731 . Google Scholar CrossRef Search ADS Jebreili M , Neshat H , Seyyedrasouli A , Ghojazade M , Hosseini MB , Hamishehkar H . 2015 . Comparison of breastmilk odor and vanilla odor on mitigating premature infants’ response to pain during and after venipuncture . Breastfeed Med . 10 : 362 – 365 . Google Scholar CrossRef Search ADS PubMed Johnson EW , Eller PM , Jafek BW . 1995 . Distribution of OMP-, PGP 9.5- and CaBP-like immunoreactive chemoreceptor neurons in the developing human olfactory epithelium . Anat Embryol (Berl) . 191 : 311 – 317 . Google Scholar CrossRef Search ADS PubMed Kobal G , Hummel T , Sekinger B , Barz S , Roscher S , Wolf S . 1996 . “Sniffin’ sticks”: screening of olfactory performance . Rhinology . 34 : 222 – 226 . Google Scholar PubMed Marlier L , Gaugler C , Messer J . 2005 . Olfactory stimulation prevents apnea in premature newborns . Pediatrics . 115 : 83 – 88 . Google Scholar CrossRef Search ADS PubMed Mennella JA , Beauchamp GK . 1996 . The human infants’ response to vanilla flavors in mother’s milk and formula . Infant Behav Dev . 19 : 13 – 19 . Google Scholar CrossRef Search ADS Mennella JA , Jagnow CP , Beauchamp GK . 2001 . Prenatal and postnatal flavor learning by human infants . Pediatrics . 107 : E88 . Google Scholar CrossRef Search ADS PubMed Russell MJ . 1976 . Human olfactory communication . Nature . 260 : 520 – 522 . Google Scholar CrossRef Search ADS PubMed Schaal B , Hummel T , Soussignan R . 2004 . Olfaction in the fetal and premature infant: functional status and clinical implications . Clin Perinatol . 31 : 261 – 85 , vi. Google Scholar CrossRef Search ADS PubMed Schaal B , Marlier L , Soussignan R . 2000 . Human foetuses learn odours from their pregnant mother’s diet . Chem Senses . 25 : 729 – 737 . Google Scholar CrossRef Search ADS PubMed Sinha AK , Sharma UK , Sharma N . 2008 . A comprehensive review on vanilla flavor: extraction, isolation and quantification of vanillin and others constituents . Int J Food Sci Nutr . 59 : 299 – 326 . doi:10.1080/09687630701539350 Google Scholar CrossRef Search ADS PubMed Soussignan R , Schaal B , Marlier L , Jiang T . 1997 . Facial and autonomic responses to biological and artificial olfactory stimuli in human neonates: re-examining early hedonic discrimination of odors . Physiol Behav . 62 : 745 – 758 . Google Scholar CrossRef Search ADS PubMed Sullivan RM , Taborsky-Barba S , Mendoza R , Itano A , Leon M , Cotman CW , Payne TF , Lott I . 1991 . Olfactory classical conditioning in neonates . Pediatrics . 87 : 511 – 518 . Google Scholar PubMed Varendi H , Porter RH , Winberg J . 1994 . Does the newborn baby find the nipple by smell ? Lancet . 344 : 989 – 990 . Google Scholar CrossRef Search ADS PubMed Yildiz, A, Arikan, D, Gözüm, S, Taştekın, A, Budancamanak, İ. 2011. The effect of the odor of breast milk on the time needed for transition from gavage to total oral feeding in preterm infants: the odor of breast milk. J Nurs Scholarsh. 43:265–273. doi:10.1111/j.1547-5069.2011.01410.x © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

Chemical SensesOxford University Press

Published: Sep 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off