Microbiological Effects of Virgin Coconut Oil Pulling in Comparison with Palm Oil Pulling as an Adjunctive Oral Hygiene Care for Patients with Gingival Inflammation: A Randomized Controlled Clinical Trial.

OBJECTIVES: Virgin coconut oil (VCO) pulling has antimicrobial activity and has been promoted as beneficial to oral health; however, limited information exists on its clinical effectiveness. This study aimed to compare the microbiological effects of VCO with palm oil (PO) pulling when used as an adjunctive oral hygiene care.
MATERIALS AND METHODS: Thirty-six volunteers with gingival inflammation were randomly assigned to start with (1) VCO (test) and (2) PO (control) interventions in a crossover design. Oil pulling was performed for 28 days adjunctively to oral hygiene routine. After a 21-day wash-out period, the participants switched the oil type and restarted the protocol. Plaque samples were collected for microbial culture at baseline, after the first oil pulling period, after washout, and after the second oil pulling period. The total, aerobic, and anaerobic bacteria and Mutans streptococci (MS) counts were recorded. The mean differences between VCO and PO were compared by paired t-test.
RESULTS: The number of total, aerobic, or anaerobic bacteria after 28 days of oil pulling was not significantly different from baseline in both PO and VCO groups. However, PO pulling demonstrated a significant reduction from baseline of MS count (P = 0.010), while VCO pulling showed no significant reduction. There was no statistically significant difference in the mean changes of any microbiological parameters between the two treatments.
CONCLUSIONS: VCO pulling did not show statistically significant superior benefit against plaque bacteria over PO pulling. Using PO pulling as an adjunctive oral hygiene care may reduce the number of MS, but this requires further investigations. Copyright:

INTRODUCTION

Oil pulling is a practice of swishing oil in the mouth that originated from Ayurveda, a traditional holistic medicine in India for 3000–5000 years. Recently, oil pulling has been popularized for various health benefits as complementary medicine.[1] It was claimed that oil pulling helps to prevent oral diseases and to treat systemic problems by pulling toxin and germs out of the bloodstream through the mucous membrane.[2] These claims of tempting benefits drive oil pulling's popularity as widely seen in various media; however, scientific evidence that supports or contradicts such claims is still lacking.[34] For oral health, a few studies reported a significant reduction in plaque index, gingival index (GI), modified GI scores, and levels of Streptococcus mutans after coconut oil pulling.[567891011] Coconut oil and palm oil (PO) have long been used as common cooking oil and are commercially available worldwide. Nowadays, virgin coconut oil (VCO) has become popular for health promotion. Unlike regular coconut oil, the wet extraction process enables VCO to be extracted directly from coconut milk under a controlled temperature.[12] VCO and PO are mainly composed of triglycerides,[13] which can be hydrolyzed into free fatty acid and monoglycerides by lipolysis activity of the saliva.[14] The predominant fatty acid composition of VCO is lauric acid (55.4%–59.1%), while it is <0.8%–1.1% in PO.[15] Lauric acid, monolaurin, and other ester derivatives in VCO may play important roles in antimicrobial activity. The major antimicrobial mechanisms proposed include cell membrane destruction by physicochemical processes in Gram-positive bacteria and lipid-coated viruses and inhibition of microbial signal transduction and transcription.[16]

Oral microorganisms in dental biofilms are responsible for major oral health problems. Gingival inflammation results from interactions between host responses and biofilm microorganisms, modified by local and systemic factors. The initiation and progression of dental caries are primarily associated with Mutans streptococci (MS) and lactobacilli.[17] A previous study showed that VCO has antimicrobial activity against S. mutans and Candida albicans biofilm in vitro;[18] however, it is not clear if it has antimicrobial effects in vivo. In contrast, PO does not contain fatty acids with antimicrobial properties. However, the physical reaction of oil pulling might have antimicrobial effects. Therefore, to evaluate the antimicrobial effects of oil pulling with VCO, which has antimicrobial ingredients, we chose to use PO as a comparison group.

Hence, the primary aim of this study is to compare the effects of VCO pulling with PO on total, aerobic, and anaerobic bacteria and MS when used as an adjunctive oral hygiene care on patients with gingival inflammation.

MATERIALS AND METHODS

Coconut (Cocos nucifera) oil used in this study was virgin cold-pressed coconut oil (Parisut®, Mada Miracle Co.,LTD., Bangkok, Thailand). Palm olein (Elaeis guineensis) oil was chosen as a control (Oleen® palm oil, Oleen Co.,LTD., Samut Sakorn, Thailand) because it has a similar level of viscosity to VCO but does not contain the antimicrobial ingredient. The oil samples were characterized by gas chromatograph/mass spectrometer/mass spectrometer using the in-house method modified from Lepage and Roy[19] at the university's Halal Science Center.

This randomized crossover clinical study (Thai Clinical Trial Registry No: TCTR20180515003) was approved by the Human Research Ethics Committee of the Faculty of Dentistry, Chulalongkorn University (HRE-DCU 2018-007) and was conducted in accordance with the Helsinki Declaration as revised in 2013. Before recruitment, all subjects were diagnosed with gingivitis (probing depth ≤3 mm) by experienced periodontists who were not involved in this study. The inclusion criteria were adult subjects who had gingival inflammation as assessed by GI ≥1[20] and agreed to refrain from any dental treatment during the study period. The GI was recorded at teeth 3, 7, 12, 19, 23, and 28, by probing at four sites/tooth (mesiobuccal/labial, mid-buccal/labial, distobuccal/labial, and mid-lingual/palatal) using a UNC-15 periodontal probe. The GI of each subject was calculated by dividing the sum of GI scores from each tooth by the number of teeth examined.

The exclusion criteria were subjects who are allergic to oil, undergoing orthodontic treatment, smokers, using mouthwash, or have a history of systemic diseases. The volunteers gave informed consent before enrollment.

The sample size was calculated based on data from a previous study by Anand et al.[21] Thirty-six participants were recruited by convenient sampling and randomly assigned into two groups using the block randomization method: Group 1 started with VCO pulling (test) and Group 2 started with PO pulling (control). Demographic data and patient history on previous periodontal treatment, medical history, and oral hygiene practice were obtained at the first data collection by interviews. For oil pulling, all participants were instructed to rinse with 10 mL of oil using the provided measuring cup, swish the oil around in the oral cavity for 8 min, and then spit out the liquid.[22] The volunteers were asked to perform oil pulling daily in the morning after their routine oral hygiene care for 28 days, followed by a 21-day washout period. Subsequently, the participants switched to the other oil type and started the oil pulling protocol again for 28 days [Figure 1]. Compliance was ensured by checking the content in the returned bottles and interviewing the participants.

Figure 1

Flowchart of the study design n – number of participants

This study was single blinded to reduce biases carried by the examiner. The plaque sample was collected by one calibrated examiner who was blinded to the status of the participants. The intra-examiner calibration was performed by examining three subjects twice in 2-h interval before the first assessment. The intraclass correlation coefficient was acceptable at >0.75.[23]

The plaque samples were collected at baseline before the start of the intervention, day 28 of the first intervention period, baseline after washout period, and day 28 of the second intervention period [Figure 1]. Plaque samples were collected from all surfaces of teeth 5, 9, 14, 21, 25, and 30 using a sterile curette. The samples from each subject were pooled and kept in an Eppendorf tube filled with 1 mL of reduced transport fluid (RTF) in an ice container. RTF was prepared as described previously,[24] kept in the 5% carbon dioxide incubator and exposed to the anaerobic atmosphere for 48 h. Finally, dithiothreitol was added within 1 h before usage.

The plaque samples were dispersed and serially diluted with RTF and spread on brain–heart infused (BHI) agar (HiMedia®, Mumbai, India) and on mitis salivarius bacitracin (MSB) agar (Difco®, Franklin Lakes, NJ, USA) using glass bead method. Then, the samples were incubated in different conditions to evaluate three groups of microorganisms: (1) aerobic microorganisms – BHI agar in 37°C aerobic incubators for 2 days, (2) anaerobic microorganisms – BHI agar in anaerobic condition (anaerobic jar) for 7 days, and (3) MS-MSB agar in 5% CO2 incubator for 2 days.

The number of aerobic, anaerobic microorganisms, and MS was measured by counting the number of colonies on the agar using a colony counter and multiplied by the dilution factor. The sum of aerobic and anaerobic microorganism colony counts was considered as total microorganisms.

The colony-forming units were log transformed (Log10CFU) for further analyses. Baseline data comparisons between the VCO group and PO group were done using Wilcoxon signed-rank test for skewed data (total microorganisms and aerobic microorganisms), while paired t-test was used for the analysis of normally distributed data (MS and anaerobic microorganisms).

The means of values at baseline were compared to those after intervention (within-group comparisons) using paired t-test for MS in VCO group while using Wilcoxon-signed rank test for aerobic, anaerobic, and total microorganisms in the coconut oil group, and MS, aerobic, anaerobic, and total microorganisms in PO group.

The mean changes of values (before–after intervention) between the two treatment groups were compared using paired t-test for all parameters. P < 0.05 was considered statistically significant. The SPSS statistics (SPSS version 22.0, SPSS Inc., IL, USA) software was used for all statistical analyses.

RESULTS

All 36 volunteers aged 19–29 years old participated throughout the study period. Subjects’ characteristics are presented in Table 1. The means of baseline log10 CFU of microorganisms are demonstrated in Table 2. There was no statistically significant difference in all parameters at baseline between the VCO and PO treatment groups.

Table 1

Characteristics of study participants

Variables	Mean±SD or n (%)
Total number	36
Age (years)	23.4±3.00
Female	25 (69.4)
GI score	1.50±0.14
Brushing technique
Scrub	3 (8.3)
Modified bass	32 (88.9)
Roll	1 (2.8)
Brushing frequency (times/days)
<2	1 (2.8)
2	31 (86.1)
>2	4 (11.1)
Interdental cleansing (Floss)	17 (47.2)

SD - Standard deviation; GI - Gingival index; n - Number of participants

Table 2

Comparison of baseline data and mean changes in microbial parameters between virgin coconut and palm oil groups

Parameters	Baseline	After oil pulling	Mean changes	Effects size
Aerobe (log10 CFU)
Virgin coconut oil	7.39±0.55	7.22±0.78	0.17±0.98	0.007
Palm oil	7.12±0.83	6.97±0.88	0.15±0.89
P	0.315b		0.914a
Anaerobe (log10 CFU)
Virgin coconut oil	7.16±0.83	7.24±0.89	−0.08±1.12	0.003
Palm oil	7.34±0.90	7.44±0.96	−0.09±1.24
P	0.292a		0.969a
MS (log10 CFU)
Virgin coconut oil	3.40±1.71	3.00±1.96	0.40±2.34	-0.179
Palm oil	3.18±1.66	2.32±1.94	0.85±1.99
P	0.599a		0.384a
Total microorganisms (log10 CFU)
Virgin coconut oil	14.55±1.11	14.46±1.17	0.09±1.47	0.005
Palm oil	14.46±1.25	14.41±1.30	0.05±1.43
P	0.838b		0.910a

aPaired t-test; bWilcoxon signed-rank test. P<0.05 was considered statistically significant. CFU - Colony-forming unit; MS - Mutans streptococci

Since VCO and PO are natural products whose composition could vary according to growth conditions/locations and extraction methods, we confirmed the composition of the oil used in this study by gas chromatograph/mass spectrometer/mass spectrometer [Table 3]. VCO contains 13 types of free fatty acids, and lauric acid (C12:0) was the most abundant. Meanwhile, PO contains 18 types of free fatty acids, with oleic acid (C18:1 n-9 cis) as the most abundant.

Table 3

Individual fatty acid composition of virgin coconut oil and palm oil used in this study

Fatty acid	Virgin coconut oil (%)	Palm oil (%)
Caproic acid (C6:0)	0.49	-
Caprylic acid (C8:0)	8.10	0.02
Capric acid (C10:0)	7.30	0.03
Lauric acid (C12:0)	35.59	0.48
Myristic acid (C14:0)	21.53	1.01
Pentadecylic acid (C15:0)	0.01	0.04
Palmitic acid (C16:0)	12.29	32.66
Palmitoleic acid (C16:1)	0.01	0.15
Margaric acid (C17:0)	-	0.07
Stearic acid (C18:0)	4.07	4.26
Oleic acid (C18:1 n-9 cis)	8.84	46.21
Linoleic acid (C18:2 n-6 cis)	1.68	13.51
Gamma-linolenic acid (C18:3 n-6)	-	0.04
Alpha-linolenic acid (C18:3 n-3)	-	0.20
Eicosenoic acid (C20:1 n-9)	-	0.14
Undecylic acid (C11:0)	0.02	-
Tridecylic acid (C13:0)	0.04	-

Figure 2 demonstrates the levels of aerobic microorganisms, anaerobic microorganisms, MS, and total microorganisms at baseline and after pulling with VCO and with PO. Comparing to the baseline, no statistically significant difference was observed for all parameters after oil pulling in the VCO group. However, after PO pulling, there was a significant reduction in the level of MS (P = 0.010) but not of the other microbial parameters. No statistically significant difference was observed between VCO and PO when the mean changes were compared [Table 2].

Figure 2

Mean score of microbial parameters before and after oil pulling (a) Aerobic microorganisms; (b) Anaerobe microorganisms; (c) Mutans streptococci (MS); (d) Total microorganisms. *P < 0.05 was considered statistically significant. aPaired t-test was performed. bWilcoxon signed-rank test was performed. CFU – Colony-forming unit

DISCUSSION

This randomized controlled trial aimed to examine whether VCO, which contains lauric acid as an antimicrobial active ingredient, is effective against dental plaque microorganisms in comparison to PO when used in oil pulling. Our study is a crossover randomized controlled trial. This design allowed us to compare the effects of the two oil types in the same individuals, which helps to reduce the influences from biological and behavioral variations among participants. We used PO as a comparison group because it has similar physical properties as VCO but does not contain antimicrobial ingredients, such as lauric acid.

Our results showed that there is no statistically significant difference in the mean changes of any microbiological parameters between VCO and PO pulling. Therefore, despite the differences in fatty acid composition and the presence of monolaurin, VCO did not show any superior effects to PO. Moreover, we observe statistically significant reductions in MS from baseline in PO pulling but not in VCO pulling group. Thus, PO may be beneficial when used as an adjunctive oral hygiene care by oil pulling method. However, further studies, particularly comparison of PO pulling with a placebo control, are required to confirm this result.

Interestingly, our results showed that VCO had no effect on microbial counts. On the other hand, PO pulling appeared to be effective in reducing the number of MS in dental plaque. This result is in contrast to the antimicrobial properties of coconut oil against S. mutans in vitro biofilm.[18] Previous clinical studies reported inconclusive results regarding the effect of coconut oil pulling on S. mutans from saliva and plaque samples.[891011] These studies differ from the current study in sample type, intervention method, sample collection, microbial quantification methods, and data analysis. In this study, we quantified MS by colony count, which is more specific and more quantitative than the Dentocult SM strip mutans kit used in previous studies.[1025] Importantly, we used logarithmically transformed data and geometric means for statistical analyses of microbial parameters. These are more appropriate due to the highly skewed nature of microbial count data,[26] but they were not used in previous studies. In addition, previous studies on the reduction of microbial counts by oil pulling reported rather small changes.[21272829] However, a reduction in the levels of orders of magnitude is often required to affect disease risks, such as the case of MS and dental caries risk (105 CFU/mL vs. 104 CFU/mL of salivary MS for high vs. low risk).[30] Although we observed a reduction of MS after PO pulling, further study is required to investigate the potential benefit of PO pulling in reducing caries risk.

The compositions of VCO and PO used in our study were analyzed by gas chromatograph/mass spectrometer/mass spectrometer. The result confirmed that PO does not contain any known active antimicrobial ingredient. Nevertheless, we observed significant reductions in MS levels in the PO group. We speculate that its chemical and mechanical properties may contribute to these effects. There are several hypotheses on the mechanisms of oil pulling that may be involved. First, oil pulling may enhance the formation of lipid-enriched pellicle on the tooth surface. An increase of hydrophobicity of the acquired pellicle may prevent protein adsorption, bacterial adherence, and subsequent biofilm formation on the tooth surface.[31] Recently, a direct visualization of hydrated pellicle-covered tooth samples after a 30-s oral rinsing with safflower or linseed oil showed surface-bound nano- and micro-sized lipid droplets that remained adherent for several hours.[32] Another mechanism could be the production of monoglycerides and free fatty acids with antimicrobial properties through emulsification and lipolysis of oil.[1433] For coconut oil, the byproducts from lipolysis activity, such as monolaurin and other medium-chain monoglycerides, may alter bacterial cell walls, disrupt cell membranes, and inhibit enzymes involved in energy production and nutrient transfer, leading to bacterial cell death.[34] However, a recent study showed that the level of lipolysis activity in the oral cavity varies among individuals, and the source of lipase enzyme is still inconclusive.[14] This may partly explain the variation in the antimicrobial effects of VCO pulling in the study population and the statistically insignificant changes in microbial parameters in this study. Another antimicrobial mechanism could be the inhibition of bacterial adhesion and plaque co-aggregation because of the oil viscosity and hydrophobicity.[3536] The higher viscosity and hydrophobicity of PO than that of VCO may also partly explain the effect of PO in our study.[3738]

While chlorhexidine gluconate antiseptic mouthwash is highly effective in reducing dental plaque, several adverse effects are observed for long-term administration.[39] Due to the potentially low adverse effects of natural remedies, oil pulling has been promoted as an adjunctive oral hygiene practice. The participants in this study reported no adverse effects during the study period.

This study carries certain limitations. First, despite a strict protocol given to the subjects, it was not possible to ensure their compliance during routine oil pulling practices. Since the oil-pulling procedure requires the participants to hold and swish the oil in their mouth for a rather long period of time, this could negatively affect their compliance. The time required for oil-pulling reported in most studies ranges from 8 to 20 min;[8922272835] thus, we chose to use the shortest time (8 min).[22] The participants reported 81.9% and 86.6% compliance for PO and VCO groups, respectively. However, the different duration of the oil pulling procedure might affect the result and should be examined in further studies. Second, the study subjects’ condition is limited to gingival inflammation. Thus, the results may not be generalizable to cases with periodontitis. Furthermore, the reduction of MS is merely a surrogate marker, whether caries risk is reduced requires a longer study period. In addition, future studies that compare PO pulling to placebo or mechanical cleansing alone and with a larger number of participants are required to clearly show the effect of oil pulling. Within the limitations of this study, our study adds to the currently scarce evidence on the effects of VCO and PO pulling on plaque microorganisms and suggests that additional studies are needed.

CONCLUSIONS

VCO pulling did not show any superior effect over PO pulling in controlling plaque microorganisms. When used as an adjunctive oral hygiene care, PO appeared to reduce the number of Mutans streptococci from baseline. Further investigations on the effectiveness of oil pulling are needed.

Financial support and sponsorship

This study was supported by the Dental Research Fund, Dental Research Project, Faculty of Dentistry, Chulalongkorn University. All microbiological experiments were supported by the Ratchadaphiseksomphot Endowment Fund to Research unit on Oral Microbiology and Immunology.

Conflicts of interest

There are no conflicts of interest.