Diabetes & Metabolic Syndrome: Clinical Research & Reviews 15 (2021) 549e555

Contents lists available at ScienceDirect

Diabetes & Metabolic Syndrome: Clinical Research & Reviews

journal homepage: [Link]/locate/dsx

Health effects of coconut oil: Summary of evidence from systematic

reviews and meta-analysis of interventional studies
Ranil Jayawardena a, b, *, Hasinthi Swarnamali c, Priyanga Ranasinghe d, Anoop Misra e
a
Department of Physiology, Faculty of Medicine, University of Colombo, Colombo, Sri Lanka
b
Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Queensland, Australia
c
Health and Wellness Unit, Faculty of Medicine, University of Colombo, Colombo, Sri Lanka
d
Department of Pharmacology, Faculty of Medicine, University of Colombo, Colombo, Sri Lanka
e
Fortis-C-DOC Centre of Excellence for Diabetes, Metabolic Diseases and Endocrinology, Chirag Enclave, New Delhi, India

a r t i c l e i n f o a b s t r a c t

Article history: Background and aims: Systemic review (SR) and meta-analysis (MA) of interventional studies are
Received 18 February 2021 considered as the highest level of evidence for clinical decision making. Therefore, we systematically
Accepted 25 February 2021 summarized all high-quality evidence on the usage of coconut oil for health-related benefits from SRs
and MA.
Keywords: Methods: PubMed®, Web of science®, SciVerse Scopus®, and EMBASE® databases were systematically
Coconut oil
searched to select SRs and SRs with MA of interventional studies reporting health-related clinical out-
Interventional studies
comes of coconut oil. Similar studies were grouped based on their respective clinical areas. A method-
Cholesterol
Systematic review
ological quality appraisal was conducted for all included SRs and SRs with MA using the Critical Appraisal
Skin Checklist for Systematic Reviews.
Oral health Results: A total of seven papers were selected for inclusion in this review, consisting of three MA and one
SR on cardio-metabolic health, one SR on oral health, and one SR and one MA each on skin health.
Coconut oil significantly increases serum total cholesterol, low-density- and high-density- lipoprotein
cholesterol levels compared to poly- and mono-unsaturated oils. Limited studies showed that topical use
of coconut oil helps in the prevention and treatment of atopic dermatitis and oil pulling for the pre-
vention of dental caries. All four studies on cardiometabolic health and the SR on oral health had a high
score in the quality assessment, SR with MA on skin health fulfilled high-quality scoring whereas the SR
on the same topic had a low-quality scoring.
Conclusions: In summary, consistent and strong evidence shows that coconut oil has an adverse effect on
the lipids parameters associated with cardio-metabolic health, with limited studies to conclude the ef-
fects of atopic dermatitis and oil pulling.
© 2021 Published by Elsevier Ltd on behalf of Diabetes India.

1. Introduction cholesterol-lowering, reduction of the risk of cardiovascular dis-

eases, weight loss, improvement of cognitive functions, irritable
In recent years, coconut oil the main product from coconut palm bowel syndrome, thyroid conditions and diabetes, enhancing the
tree [1] has gained wider attention. Many health professionals and immune system, and promoting wound healing [3e5]. Coconut oil
other authorities have endorsed the use of coconut oil as a cooking has several unique characteristics that may be responsible for these
medium in substitution to other vegetable oils and as a supple- health benefits. Coconut oil contains nearly 50% of medium-chain
mentary ingredient [1,2]. Scientific and lay literature both promotes fatty acids (MCFA), which are believed to be converted to ketone
the consumption of coconut oil based on assumptions that it has bodies [6] that pass through the blood-brain barrier and provide
many health benefits. These postulated benefits include energy for glucose depleted brain [7], with a possible therapeutic
role in neurodegenerative diseases like Alzheimer’s and Parkinson’s
diseases [8]. Moreover, MCFA oxidized efficiently, deposited less in
* Corresponding author. Department of Physiology, Faculty of Medicine, Univer- adipose tissue, and increase diet-induced thermogenesis and en-
sity of Colombo, Colombo, Sri Lanka. ergy expenditure than long-chain fatty acids (LCFA) [9].
E-mail address: ranil@[Link] (R. Jayawardena).

[Link]
1871-4021/© 2021 Published by Elsevier Ltd on behalf of Diabetes India.
R. Jayawardena, H. Swarnamali, P. Ranasinghe et al. Diabetes & Metabolic Syndrome: Clinical Research & Reviews 15 (2021) 549e555

2. Methods
Abbreviation
A systematic review of SRs and SRs with MA on the effects of
CVD cardiovascular diseases coconut oil on clinical outcomes was performed in accordance with
HDL-C: high-density lipoprotein cholesterol the Preferred Reporting Items for Systematic Review and Meta-
HABSI hospital-acquired bloodstream infections Analysis (PRISMA) guidelines (Supplementary Material 1) [22].
LCFA long-chain fatty acids
LDL-C: low-density lipoprotein cholesterol 2.1. Search strategy
MA Mata-analysis
MCFA medium-chain fatty acids A comprehensive search of the medical literature was con-
MCT medium-chain triglycerides ducted in a stepwise process in the following databases; PubMe-
MD mean difference d®(U.S. National Library of Medicine, USA), Web of science® [v.5.4]
MUFA monounsaturated fatty acid (Thomson Reuters, USA), SciVerse Scopus® (Elsevier Properties S.A,
PUFA polyunsaturated fatty acid USA), and EMBASE® (Elsevier Properties, Netherlands) for studies
RR relative risk published before December 1, 2020. In the first stage, a combina-
SR Systemic review tion of keywords was used to identify potentially relevant studies in
TC total cholesterol the above-specified databases (Supplementary Material 2). Search
TG triglycerides limits were species (“humans”), and language (“English”). Two re-
viewers (HS and RJ) performed study selection independently.
In the second stage, the total hits obtained from searching the
above databases were pooled together and duplicate articles were
Furthermore, MCFA has fewer calories per gram compared to removed. Retrieved articles were screened by reading the article
polyunsaturated fatty acid (PUFA), suggesting beneficial effects on ‘title’ in the third stage and thereafter the ‘abstracts’ in stage four.
weight loss. Other than MCFA; polyphenols, antioxidants, vitamin Studies not satisfying the inclusion criteria (given below) were
E, sterols, and phospholipids contained in coconut oil are also excluded. The remaining articles were screened at the final stage by
attributed to the various postulated health effects [10]. reading the complete manuscript and those not satisfying the
Although numerous health benefits have been documented, the eligibility criteria were excluded. To obtain additional data, a
highest-quality evidence from SRs and SR with MA on the effect on manual search was conducted using the reference lists from the
coconut oil and cardio-metabolic parameters shows a negative included articles. Any possible forward citations of the studies
impact on health [11,12]. Moreover, there are no strong physiological retrieved during the literature search were traced and screened for
explanations of some mechanisms for the health benefits of coconut possible inclusion. This searching process was carried out inde-
oil. For example, claims relating to coconut oil for the treatment of pendently by two reviewers (HS and RJ) and the final group of ar-
Alzheimer’s disease are based on supplementing caprylic acid (C8) ticles to be included in the review was determined after an iterative
and capric acid (C10) which results in increased plasma ketone body consensus process.
concentrations leading to positive cognitive performance [13,14].
However, in reality, C8 and C10 account for only 10% of the total fat 2.2. Inclusion and exclusion criteria
derived from coconut oil [6]. Therefore, it is very unlikely to get a
similar effect derived from supplementing medium-chain tri- The studies were selected according to the following Inclusion
glycerides (MCT) (C8, C10) by using coconut oil alone. Moreover, criteria: a) systematic review or systematic review with meta-
coconut oil has been described for its benefit on weight loss due to its analysis of intervention studies reporting any health effect of co-
high MCT content. However, lauric acid is the major fatty acid conut oil; and Exclusion criteria: a) studies published as original
encountered in coconut oil, and although it is considered as a MCFA studies, letters, conference abstracts, opinion papers, non-
based on biochemical properties [15], it has also been classified as systematic reviews, or books; b) if coconut oil was part of a
either MCFA or LCFA in terms of digestion and metabolism [16]. mixed intervention and if it is not possible to verify individual
Therefore, the fact that coconut oil aids in weight loss is still impact c) if the systematic review evaluated effects of mainly lauric
debatable because lauric acid does not behave similarly to other acid or MCFA/MCT oil rather than coconut oil; e) if the review was
MCT. Furthermore, various biological effects of coconut oil, such as related to the history of coconut production or use.
blood pressure and cholesterol-lowering, and potential as an Alz-
heimer’s treatment have also been attributed to its phenolic content 2.3. Data extraction and analysis
and antioxidant potential [17e19]. Nonetheless, phenols are not
considered as an antioxidant by the Food and Drug Administration Data were summarized in tables from the included studies by
agency, since although it has antioxidant activity in-vitro, similar one reviewer (HS) using a standardized form and checked for ac-
effects are unlikely to be observed in-vivo [20,21]. curacy by a second reviewer (RJ). The following information was
The bulk of the evidence often used to back the health benefits extracted from each study: (1) Details of the study (lead author,
of coconut oil is mostly based on observational studies with inferior country, year of publication), (2) the number of interventions under
methodology. The beneficial effect of coconut oil on health is the SR and MA, (3) parameters evaluated (4) Findings of SR or SR
exciting but not well explained. At the same time, most of the with MA (summary of MA including mean difference [MD]or
beneficial properties of coconut oil have been derived by animal relative risk [RR]and its p-value), (5) Statistical heterogeneity of the
studies or small isolated clinical trials. Therefore, the examination MA or estimates of the proportion of variance reflecting true dif-
of prevailing reviews on coconut oil consumption and the scientific ferences in effect size (I [2] statistics), (6) Any presented measure of
evidence behind its health utility is important. Systemic review publication bias (Funnel plots or Egger’s regression test). The out-
(SR) and meta-analysis (MA) of international studies are considered comes were categorized as follows; cardio-metabolic health, oral
as the highest level evidence for clinical decisions. Therefore, we health, and skin health. Any discrepancies in the data extracted
aimed to summarize all high-quality evidence arising from SR and were rechecked and resolved by discussion with the third reviewer
MA on coconut oil in this review. (PR).
550
R. Jayawardena, H. Swarnamali, P. Ranasinghe et al. Diabetes & Metabolic Syndrome: Clinical Research & Reviews 15 (2021) 549e555

2.4. Data analysis All 4 studies on cardio-metabolic health [11,12,24,25] were

deemed to be of high-quality, scoring between 8 and 10 (Neela-
A methodological quality appraisal was conducted for all SR and kantan et al. [12]e 10, Teng et al. [25]- 10, Jayawardena et al. [11] e
SR with MA using the Critical Appraisal Checklist for systematic 9 and Eyres et al. [24] e 8). SR on oral health [26] fulfilled the
Reviews, developed by the Umbrella Review Methodology Working criteria for high-quality obtaining a score of 10. SR with MA on skin
Group [23]. This check list consists of 10 items; each item within the health [27]satisfied high-quality scoring (10 scores) whereas the
instrument can receive 1 point, for an overall quality score that SR28alone had a low-quality scoring (3 scores). Scores for all
could range from 0 to 10. Meta-analyses with quality scores ranging selected reviews under each point in the critical appraisal checklist
from 0 to 4 were labeled as low quality, those with scores between for systematic reviews and research syntheses are summarized in
5 and 7 as medium quality, and those with scores of 8e10 as high Supplementary Material 3.
quality. Because this is a descriptive summary review of meta-
analyses, no statistical analyses were performed. 3.1. Cardio-metabolic health

3. Results A systematic review by Eyres et al. (2016) [24] examined seven

interventional (sample size range from 9 to 83) reporting on co-
Literature search was done according to the above search conut oil intake and lipid profiles, which comprised of three ran-
criteria and the search strategy is summarized in Fig. 1. The data- domized cross-over, one randomized parallel, and three sequential
base search identified 805 articles and 535 studies remained after feeding trials. The evidence from the interventional studies showed
removing duplicates. Screening by title and abstract reduced the that coconut oil intake led to increased levels of low-density lipo-
number to 25. A total of seven papers were selected for inclusion in protein cholesterol (LDL-C), high-density lipoprotein cholesterol
this review [11,12,24e28]. Four studies on cardio metabolic health (HDL-C), and total cholesterol (TC) levels, compared to other
(three with MA and one Stallone) [11,12,24,25] (Table 1a), one SR on vegetable oils. Among the studies analyzed, coconut oil intake
oral health [26] (Table 1b), and two articles were on skin health increased LDL-C in six studies, TC in seven and HDL-C in five. No
(one SR on preterm/low birth weight infants [27]and other SR was a discernible effects were observed regarding serum triglycerides
MA on neurosis patients [28]) (Table 1c). (TG).

Fig. 1. Flow diagram of the literature search process.

(PICO, participants, intervention, comparison, and outcome).

551
R. Jayawardena, H. Swarnamali, P. Ranasinghe et al. Diabetes & Metabolic Syndrome: Clinical Research & Reviews 15 (2021) 549e555

Table 1a
Effect of coconut oil on cardio-metabolic health.

Authors(ref); No. of studies Parameters evaluated Main Findings of SR or SR with MA I2 Publication

Year of in SR:MA bias
publication

Eyres et al. 7:0 TC, LDL, HDL, TG, VLDL, LDL:HDL, TG:HDL, TC:HDL, Apo A1, S[ TC in 7 trials cf. cis-USF. NR NR
[24]; Apo B S[ LDL in 6 trials cf. cis-USF.
2016 S[ HDL in 5 trials cf. cis-USF.
S[TG in 2 trials cf. corn oil.
S[LDL:HDL in 1 trial cf. corn oil.
SYTC: HDL in 1 trial cf. olive oil.
Teng et al. 18 : 12 TC, LDL, HDL, TG S[LDL (0.26 mg/dL; p < 0.05) cf. MUFA & PUFA plant oils. 60% P ¼ 0.789a
[25]; S[HDL (0.57 mg/dL; p < 0.05) cf. MUFA & PUFA plant oils. 7% P ¼ 0.283a
2019 S[HDL (0.33 mg/dL; p < 0.05) cf. animal fats. 0% NR
SYLDL (
0.37 mg/dL; p < 0.05 cf. animal fats. 48% NR
S[LDL (0.43 mg/dL; p < 0.05) cf. PUFA plant oils. 82% NR
S[TG (0.31 mg/dL; p < 0.05) cf. PUFA plant oils. 32% NR
Jayawardena 23: 20 TC, LDL, HDL, TG, VLDL, LDL:HDL, TG:HDL, FPG, HbA1c, BW, S[ TC (15.42 mg/dL; p < 0.001) cf. other oils and fats 83% NR
et al. [11]; BMI, WC, Apo A1, Apo B, hs-CRP S[ LDL (10.14 mg/dL; p < 0.001) cf. other oils and fats 73%
2020 S[ HDL (2.61 mg/; p ¼ 0.002) cf. other oils and fats 34%
S[ Apo A1 (0.14 mg/dL; p < 0.001) cf. other oils and fats 0%
SY TC:HDL (0.2; p ¼ 0.004) cf. other oils and fats 7%
SYHbA1c (0.39%; p < 0.001) cf. other oils and fats 30%
S[TC (50.82 mg/dL; p < 0.001) cf. corn oil. 0%
S[ TC (25.86 mg/dL; p ¼ 0.007) cf. palm oil. 71%
S[ TC (14.40 mg/dL; p ¼ 0.02) cf. soybean oil. 68%
S[ TC (11.50 mg/dL; p ¼ 0.05) cf. safflower oil 54%
S[ TC (12.22 mg/dL; p ¼ 0.20) cf. olive oil. 77%
S[ LDL (20.84 mg/dL; p ¼ 0.008) cf. palm oil. 55%
S[ LDL (11.50 mg/dL; p ¼ 0.03) cf. soybean oil. 55%
S[ LDL (8.05 mg/dL; p ¼ 0.04) cf. safflower oil. 31%
S[ LDL (7.73 mg/dL; p ¼ 0.37) cf. olive oil. 76%
S[ HDL (2.61 mg/dL; p ¼ 0.002) cf. butter. 4%
SY LDL (
14.90 mg/dL; p < 0.001) cf. butter. 0%
Neelakantan 16 : 16 TC, LDL, HDL, TG, BW, WC, %BF, FPG, CRP S[TC (14.69 mg/dL; p < 0.05) cf. non-tropical vegetable oils. 91% P ¼ 0.57a
et al. [12]; S[LDL (10.47 mg/dL; p < 0.05) cf. non-tropical vegetable 84% P ¼ 0.36a
2020 oils. 72% P ¼ 0.002a
S[HDL (4.00 mg/dL; p < 0.05) cf. non-tropical vegetable oils. 79% NR
S[TC (25.57 mg/dL; p < 0.05) cf. palm oil. 67% NR
S[LDL (20.50 mg/dL; p < 0.05) cf. palm oil. 29% NR
S[ HDL (2.83 mg/dL; p < 0.05) cf. palm oil.

ApoA: apoprotein A; ApoB: Apoprotein; BF: body fat; BMI: body mass index; cf: compare to; CO: coconut oil; CRP: c-reactive protein; HDL: high density lipoproteins
cholesterol; I [2]: heterogeneity; LDL: low density lipoprotein cholesterol; MA: meta analysis; MUFA: mono unsaturated fatty acids; NR: not reported; PUFA: polyunsaturated
fatty acids; RBD: refined, bleached and deodorized; S: significant; SR: systematic review; TC: total cholesterol; TG: triglycerides; USF: unsaturated fats; WC: waist
circumference.
a
Egger’s regression test.

Table 1b
Effect of coconut oil on oral health.

Authors(ref); Year No. of studies in Parameters evaluated Main Findings of SR and/or MA I2 Publication
SR:MA bias

Woolley et al. [26]; 4:0 Salivary BC count, Plaque-, Gingival-, Stain-index score, Bleeding SY Salivary BC count (p  0.03) cf. NR NR
2020 on probing, distilled water.
SY Plaque index score (p ¼ 0.001) cf.
distilled water.
SY Staining (p ¼ 0.0002) cf. distilled
water.

BC: bacterial colony; cf: compare to; I [2]: heterogeneity; MA: meta-analysis; NR: nit reported; SR: systematic review.

Teng et al. (2019) [25] has done a systematic review and meta- TG (0.31 mg/dL, p < 0.05) when compared with PUFA-rich plant oils
analysis which reviewed 18 clinical trials, of which 12 were used (soybean and safflower [high linoleum variety] oils), but it had no
for meta-analysis. Of those 18 clinical trials, 11 were randomized significant effect on LDL-C and TG when compared with MUFA-rich
trials whereas the other seven were non-randomized clinical trials. plant oils (olive, peanut, canola, and safflower [high oleic variety]
The sample size varied from 9 to 200 in included studies. They have oils).
found that coconut oil significantly increased LDL-C (0.26 mg/do, Systematic review and meta-analysis by Jayawardena et al.
p < 0.05) and HDL-C (0.57 mg/dL, p < 0.05) compared with plant (2020) [11] comprised of 23 interventional studies, of which 20
oils, while significantly increasing HDL-C (0.33 mg/dL, p < 0.05) and studies have been used for meta-analysis. From 23 studies, 13 were
decreasing LDL-C (0.37 mg/dL, p < 0.05) compared with animal fats. randomized cross-over, seven were randomized parallel and three
Additionally, they have performed a sub-group analysis in com- were sequential feeding trials. The sample size varied from 9 to 190
parison to MUFA-rich and PUFA-rich plant oils. Results showed that in included studies. This also concluded that there were detri-
coconut oil significantly increased LDL-C (0.43 mg/dL, p < 0.05) and mental effects on cardio-metabolic health by coconut oil compared

552
R. Jayawardena, H. Swarnamali, P. Ranasinghe et al. Diabetes & Metabolic Syndrome: Clinical Research & Reviews 15 (2021) 549e555

with corn oil, palm oil, soybean oil and safflower oil due to signif-

NR

NR
icant increase in TC (50.82 mg/dL, p < 0.001; 25.86 mg/dL,

Publication
p ¼ 0.007; 14.40 mg/dL, p ¼ 0.02; 11.50 mg/dL, p ¼ 0.05; 12.22 mg/
dL, p ¼ 0.20 respective oils). The LDL-C was significantly increased

bias
by coconut oil compared with palm oil (20.84 mg/dL, p ¼ 0.008),
soybean oil (11.50 mg/dL, p ¼ 0.03), safflower oil (8.05 mg/dL,
p ¼ 0.04) and olive oil (7.73 mg/dL, p ¼ 0.37) [11]. There was a

cf: compare to; HABSI: hospital-acquired blood stream infections; HC: head circumference; I [2]: heterogeneity; NR: not reported; SCORAD: scoring atopic dermatitis; TEWL: transepidermal water loss.
NR positive effect of coconut oil when compared with butter, as co-

NR
0%
I2

conut oil raised HDL-C (2.61 mg/dL, p ¼ 0.002) and lowered LDL-C
(14.90 mg/dL, p < 0.001) respectively.
Neelakantan and co-workers (2020) [12] also conducted a sys-
tematic review and meta-analysis of coconut oil on cardio-
SY Staphylococcus aureus colonization cf. olive & mineral oil.

S[ Length gain velocity (p < 0.05) cf. mineral oil/placebo metabolic risk factors which included 16 interventional studies
involving 730 participants.12Similar to earlier findings, coconut oil
consumption significantly increased TC (14.69 mg/dL, p < 0.05) and
LDL-C (10.47 mg/dL, p < 0.05) compared with non-tropical vege-
SY HABSI (RR ¼ 0.35; p ¼ 0.001) cf. placebo.

table oils (soybean, olive, safflower, and canola oils) although it

slightly increased HDL-C (4.00 mg/dL, p < 0.05). Besides, they have
shown that palm oil (another tropical oil with z50% saturated fat
S[ Skin capacitance cf. mineral oil
S Improved xerosis cf. mineral oil.

vs z 90% saturated fat in coconut oil) was also better than coconut
SYSCORAD index cf. mineral oil.
Main Findings of SR and/or MA

oil, as coconut oil significantly increased TC (25.57 mg/dL, p < 0.05)

and LDL-C (20.50 mg/dL, p < 0.05) compared with palm oil.
SYTEWL cf. control.

SYTEWL cf. control.

3.2. Oral health

A systematic review (2020) of four randomized controlled trials

(sample size range from 30 to 60) reported that oil pulling with
coconut oil has beneficial effects on improving oral health and
dental hygiene as it significantly reduced salivary bacterial colony
count (p  0.03) and plaque index score (p ¼ 0.001) compared to
distilled water and mouth rinse [26]. One study also demonstrated
HABSI, z scores for Wt, length and HC, TEWL.

a significant difference in staining (p ¼ 0.0002) compared to using

Chlorhexidine. Therefore, this review showed the clinical signifi-
cance of oil pulling with coconut oil could be used as an adjunct to
normal preventative regimes to improve oral health and dental
Sebumeter assessment scores,

hygiene.
Parameters evaluated

TEWL, SCORAD index

Corneometer and

3.3. Skin health

There were two reviews on skin health comprising of one SR

with MA and one SR alone. The SR with MA comprised of seven
randomized clinical trials (sample size range from 34 to 117) [27]
while the SR without MA contained four randomized clinical trials
(sample size range from 64 to 258) [28]. SR without MA done by
Karagounis et al. (2018) found that antimicrobial properties of co-
No. of studies in SR:MA

conut oil may account for the observed positive effects on atopic
dermatitis and beneficial complementary treatment for xerosis
because the topical application of coconut oil significantly reduced
transepidermal water loss, scoring atopic dermatitis index, and
Staphylococcus aureus colonization, while significantly increased
skin capacitance and improved xerosis [28].
4:0

7:2

Findings from SR with MA by Pupala et al. (2019) were derived

Effect of coconut oil on skin health.

from preterm and low birth weight (LBW) infants (n ¼ 727 infants).
This revealed that coconut oil significantly reduced hospital-
acquired bloodstream infections (HABSI) (p ¼ 0.001), also signifi-
Karagounis et al. [28];

cantly increased length gain velocity (p < 0.05) of preterm and LBW
Pupala et al. [27];

infants [27]. Finally, this review concluded that topical application

of coconut oil on skin has a beneficial effect on decreased water
Authors(ref);

loss, better growth, and skin condition, reducing the risk of HABSI,
improving the skin integrity and skin nutrition of preterm infants.
Table 1c

2018

2019
Year

There were no significant adverse effects associated with the

topical application of coconut oil.
553
R. Jayawardena, H. Swarnamali, P. Ranasinghe et al. Diabetes & Metabolic Syndrome: Clinical Research & Reviews 15 (2021) 549e555

4. Discussion and Candida albicans [36]. Methodological quality of this SR fulfilled

high-quality scoring. However, this SR highlighted the absence of
This summary review was based on the highest level of evidence high-quality evidence in the literature subjected to bias. Conse-
that aimed to provide insights into the impact of coconut oil on quently, it is therefore difficult to determine whether oil pulling
various health effects. As SR and MA employ a structured process to with coconut oil has an actual beneficial effect. Furthermore, the
comprehensively identify all relevant studies and synthesize the comparison was made with distilled water as opposed to the use of
available evidence on a particular question, a summary of those other oils.
studies assimilate the vast amount of high-level evidence of the There were one SR and one SR with MA for evaluating the effect
effect of coconut oil on all health outcomes. There are several of coconut oil on skin health. Both of these reviews reported
narrative reviews on the beneficial effect of coconut oil on human improved skin wellbeing through various parameters such as
health; however, to the best of our knowledge, this is the first hospital-acquired bloodstream infections, transepidermal water
systematic review to assess the pooled effect of all existing the loss and length gain velocity of preterm infants and, improved
highest-quality evidence on coconut oil. xerosis, scoring atopic dermatitis index and Staphylococcus aureus
Concerning coconut oil on cardio-metabolic health, three of the colonization of patients with atopic dermatitis, compared to other
four reviews reported a significant increase in TC concentration in mineral oils. Emollients have the potential to enhance the skin
coconut oil compared to poly- or mono-unsaturated oils or palm barrier function. Various emollients including natural oils, coconut
oil. Of the four reviews that investigated coconut oil effects on LDL- oil have shown the superior potential to improve clinical outcomes
C and HDL-C concentration, all of them reported a statistically because of their high content of MCFAs, especially lauric acid [37].
significant increase in both LDL-C and HDL-C compared to other Lauric acid and its ester, monolaurin, both have better anti-
vegetable oils. Additionally, MAs by Jayawardena et al. and Neela- inflammatory and antimicrobial properties compared to mineral
kantan et al. further showed this increment is much higher for LDL- oils [38e40]. Methodological quality of SR with MA satisfied high-
C than HDL-C. However, butter or animal fat which has a high quality scoring whereas SR reported low-quality scoring, but was
amount of long-chain saturated fats may have a greater negative still included in this summary review because it provided useful
impact on TC and LDL-C compared to coconut oil which has more information regarding the effectiveness of the topical application of
medium-chain saturated fats. Furthermore, two of them reported a coconut oil for skin health.
significant increase in TG concentration compared with unsatu-
rated oils. Coconut oil is made up of about 90% saturated fats and 9%
4.1. Strengths and weaknesses
unsaturated fats and composed of the fatty acids, caprylic acid (8%),
capric acid (7%), lauric acid (49%), myristic acid (8%), palmitic acid
This review has systematically summarized the current evi-
(8%), stearic acid (2%), oleic acid (6%) and 2% of linoleic acid [29].
dence for coconut oil and all health outcomes for which a previous
Therefore, the fatty acid which is high in coconut oil is lauric and it
MA and/or SR had been conducted. It used systematic methods that
is responsible for many properties of coconut oil. According to the
included a robust search strategy of four scientific literature data-
literature, lauric acid is the fatty acid that increases TC and LDL-C to
bases with independent study selection and extractions by two
the greatest extent [30e32]. However, it also has the greatest HDL-
investigators. We used standard approaches to assess the quality of
C raising effect as well [30,31]. A ratio of TC to HDL-C was examined
methods using the Critical Appraisal Checklist for Systematic Re-
in two reviews which reported a significantly lower ratio of TC to
views and quality. This score assisted us in identifying the quality of
HDL-C following a coconut oil diet compared to diets with unsat-
the review presented for each outcome, however, it does not equate
urated oils. As coconut oil has a greater HDL-C raising effect as well
to the high quality of the original studies.
other than LDL-C [30,31], decreased TC:HDL-C ratio is apparent in
A possible limitation of our review was that we did not re-
those two reviews. Past literature has shown that the TC:HDL-C
analyze any of the MA as the data needed to compute these were
ratio is an alternative marker that has been proposed to better
not generally available in the articles. We did not review the pri-
reflect CVD risk than the individual effect of increased LDL-C or
mary studies included in each of the SR and MA that would have
decreased HDL-C [33]. However, the effect on this ratio was eval-
facilitated this. There were a few SR and MA on a similar topic as
uatedin only two systematic reviews. In addition to that, the LDL-C
well as the same studies in different SR, so results will be the same.
to HDL-C ratio was evaluated in only one of the systematic reviews
We used data available in the published SR and MA and therefore
which reported a significantly higher LDL-C: HDL-C ratio after a
assumed the exposure and estimate data for component studies
coconut oil diet compared with an unsaturated oil diet. Thus, the
had been published accurately. However, SR and MA are limited by
cholesterol-raising effect of lauric acid is proportionally higher for
the quality of the primary studies. In the present review, the
LDL-C than for HDL-C. The methodological quality of all these four
included studies generally had methodological limitations,
MA and/or SR achieved a score ranging from eight to ten from the
including small sample size and short study duration.
Critical Appraisal scale for Systematic Reviews, developed by the
Umbrella Review Methodology Working Group.
There was only one SR for reporting the effect of coconut oil 5. Conclusions
when used for oil pulling to improve dental hygiene and on oral
health [26]. The procedure of oil pulling involves swishing a Based on available reviews, coconut oil consumption increases
measured volume of oil around the mouth for a while, forcing the the serum TC, LDL-C, and HDL-C levels. PUFAs and MUFAs should be
oil in between all the teeth and around the mouth [34].Examples of preferentially encouraged over coconut oil as they lower LDL-C and
organic oils that are used as oil pulling include sunflower oil, ses- increase HDL-C significantly. The effect of coconut oil on CVD risk
ame oil, and coconut oil [34].However, Coconut oil is rich in MCFA; markers other than plasma lipids is largely unreported. Limited
it is therefore unique compared to the majority of other dietary oils, evidence supports the topical use in the prevention and treatment
which are predominantly made up of LCFA. Lauric acid, the pre- of atopic dermatitis, as well as in oil pulling for the prevention of
dominant MCFA in coconut known for its antimicrobial and anti- dental caries. The available data on oral and skin health suggests
inflammatory benefits, which is the rationale for giving dental that further well-designed randomized controlled trials are
hygiene [35]. A previous in-vitro study has demonstrated the required to determine the impact of these outcomes with coconut
antimicrobial properties of coconut oil against Streptococcus mutans oil.
554
R. Jayawardena, H. Swarnamali, P. Ranasinghe et al. Diabetes & Metabolic Syndrome: Clinical Research & Reviews 15 (2021) 549e555

DECLARATIONS; authors’ contributions [16] Denke MA, Grundv SM. Comparison of effects of lauric acid and palmitic acid
on plasma lipids and Lipoproteins13vol. 56; 1992. Accessed November 19,
2018, [Link]
RJ devised the conceptual idea. HS and RJ searched databases. 5973&amp;rep¼rep1&amp;type¼pdf.
HS and RJ were involved in retrieving data. HS, RJ and PR drafted [17] Nevin KG, Rajamohan T. Beneficial effects of virgin coconut oil on lipid pa-
the manuscript. AM revised the manuscript. All authors provided rameters and in vitro LDL oxidation. Clin Biochem 2004;37(9):830e5. https://
[Link]/10.1016/[Link].2004.04.010.
critical feedback on the manuscript. All authors read and approved [18] Nurul-Iman BS, Kamisah Y, Jaarin K, Qodriyah HMS. Virgin coconut oil pre-
the final manuscript.; Ethics declarations; Not applicable. vents blood pressure elevation and improves endothelial functions in rats fed
with repeatedly heated palm oil. Evid base Compl Alternative Med 2013;2013.
[Link]
Declaration of competing interest [19] Fernando WMADB, Martins IJ, Goozee KG, Brennan CS, Jayasena V,
Martins RN. The role of dietary coconut for the prevention and treatment of
The authors declare no conflict of interest. RJ presented most of Alzheimer’s disease: potential mechanisms of action. Br J Nutr 2015;114(1):
1e14. [Link]
the materials in the Ahuja Bajaj Symposium-VI, New Delhi, India on [20] Chapter 17. Dietary antioxidants: a consideration of factors influencing re-
November 22, 2020. quirements. Accessed, [Link]
[Accessed 30 January 2021].
[21] Cory H, Passarelli S, Szeto J, Tamez M, Mattei J. The role of polyphenols in
Acknowledgments human health and Food systems: a mini-review. Front Nutr 2018;5:87.
[Link]
None. [22] Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting
systematic reviews and meta-analyses of studies that evaluate healthcare
interventions: explanation and elaboration. BMJ 2009;339. [Link]
Appendix A. Supplementary data 10.1136/bmj.b2700.
[23] Aromataris E, Fernandez R, Godfrey CM, Holly C, Khalil H, Tungpunkom P.
Supplementary data related to this article can be found at Summarizing systematic reviews. Int J Evid Base Healthc 2015;13(3):132e40.
[Link]
[Link] [24] Eyres L, Eyres MF, Chisholm A, Brown RC. Coconut oil consumption and car-
diovascular risk factors in humans. Nutr Rev 2016;74(4):267e80. https://
Funding [Link]/10.1093/nutrit/nuw002.
[25] Teng M, Zhao YJ, Khoo AL, Yeo TC, Yong QW, Lim BP. Impact of coconut oil
consumption on cardiovascular health: a systematic review and meta-Anal-
None. ysis. Nutr Rev 2020;78(3):249e59. [Link]
[26] Woolley J, Gibbons T, Patel K, Sacco R. The effect of oil pulling with coconut oil
to improve dental hygiene and oral health: a systematic review. Heliyon
References 2020;6(8):e04789. [Link]
[27] Pupala SS, Rao S, Strunk T, Patole S. Topical application of coconut oil to the
[1] Coconut oil - an overview | science direct topics. Accessed, [Link] skin of preterm infants: a systematic review. Eur J Pediatr 2019;178(9):
[Link]/topics/medicine-and-dentistry/coconut-oil. [Accessed 11 1317e24. [Link]
February 2021]. [28] Karagounis TK, Gittler JK, Rotemberg V, Morel KD. Use of "natural" oils for
[2] Da Silva Lima R, Block JM. Coconut oil: what do we really know about it so far? moisturization: review of olive, coconut, and sunflower seed oil. Pediatr
Food Qual Saf 2019;3(2):61e72. [Link] Dermatol 2019;36(1):9e15. [Link]
[3] Wallace TC. Health effects of coconut oil-A narrative review of current evi- [29] Deb Mandal M, Mandal S. Coconut (Cocos nucifera L.: arecaceae): in health
dence. J Am Coll Nutr 2019;38(2):97e107. [Link] promotion and disease prevention. Asian Pac J Trop Med 2011;4(3):241e7.
07315724.2018.1497562. [Link]
[4] Deb Mandal M, Mandal S. Coconut (Cocos nucifera L.: arecaceae): in health [30] Mensink RP, Zock PL, Kester AD, Katan MB. Effects of dietary fatty acids and
promotion and disease prevention. Asian Pac J Trop Med 2011;4(3):241e7. carbohydrates on the ratio of serum total to HDL cholesterol and on serum
[Link] lipids and apolipoproteins: a meta-analysis of 60 controlled trials. Am J Clin
[5] Deen A, Visvanathan R, Wickramarachchi D, et al. Chemical composition and Nutr 2003;77(5):1146e55. [Link]
health benefits of coconut oil: an overview. J Sci Food Agric. Published online [31] Micha R, Mozaffarian D. Saturated fat and cardiometabolic risk factors, coro-
October 29, 2020:jsfa.10870. doi:10.1002/jsfa.10870. nary heart disease, stroke, and diabetes: a fresh look at the evidence. Lipids
[6] Dayrit FM. Lauric acid is a medium-chain fatty acid. Coconut oil is a medium- 2010;45(10):893e905. [Link]
chain triglyceride, vol. 143; 2014. [32] Encyclopedia of human nutrition - second ed.. Accessed January 28, 2021.
[7] Pardridge WM. Transport of nutrients and hormones through the blood-brain [Link]
barrier. Diabetologia 1981;20(1 Supplement):246e54. [Link] caballero/978-0-08-045428-3.
10.1007/BF00254490. [33] Iggman D, Rise rus U. Role of different dietary saturated fatty acids for car-
[8] Jensen NJ, Wodschow HZ, Nilsson M, Rungby J. Effects of ketone bodies on diometabolic risk. Clin Lipidol 2011;6(2):209e23. [Link]
brain metabolism and function in neurodegenerative diseases. Int J Mol Sci clp.11.7.
2020;21(22):8767. [Link] [34] Shanbhag VKL. Oil pulling for maintaining oral hygiene e a review. J Tradit
[9] Moussavi N, Gavino V, Receveur O. Could the quality of dietary fat, and not Complement Med 2017;7(1):106e9. [Link]
just its quantity, be related to risk of obesity. Obesity 2008;16(1):7e15. [Link].2016.05.004.
[Link] [35] Peedikayil F, Sreenivasan P, Narayanan A. Effect of coconut oil in plaque
[10] Coconut oil boom. Accessed, [Link] related gingivitis - a preliminary report. Niger Med J 2015;56(2):143. https://
magazine/featured-articles/coconut-oil-boom-may-2016?SSO¼True. [Link]/10.4103/0300-1652.153406.
[Accessed 30 November 2020]. [36] Naseem M, Khiyani MF, Nauman H, Zafar MS, Shah AH, Khalil HS. Oil pulling
[11] Jayawardena R, Swarnamali H, Lanerolle P, Ranasinghe P. Effect of coconut oil and importance of traditional medicine in oral health maintenance. Int J
on cardio-metabolic risk: a systematic review and meta-analysis of inter- Health Sci 2017;11(4):65e70. Accessed, [Link]
ventional studies. Diabetes Metab Syndr Clin Res Rev 2020;14(6). https:// pubmed/29085271. [Accessed 1 February 2021].
[Link]/10.1016/[Link].2020.09.033. [37] Fatty acid composition and possible health effects of coconut constituents -
[12] Neelakantan N, Seah JYH, Van Dam RM. The effect of coconut oil consumption PubMed. Accessed, [Link] [Accessed 28
on cardiovascular risk factors: a systematic review and meta-analysis of January 2021].
clinical trials. Circulation 2020:803e14. [Link] [38] Preuss HG, Echard B, Enig M, Brook I, Elliott TB. Minimum inhibitory con-
LATIONAHA.119.043052. Published online. centrations of herbal essential oils and monolaurin for gram-positive and
[13] Reger MA, Henderson ST, Hale C, et al. Effects of b-hydroxybutyrate on gram-negative bacteria. Mol Cell Biochem 2005;272(1e2):29e34. https://
cognition in memory-impaired adults. Neurobiol Aging 2004;25(3):311e4. [Link]/10.1007/s11010-005-6604-1.
[Link] [39] Antibacterial study of the medium chain fatty acids and their 1-
[14] Taylor MK, Sullivan DK, Mahnken JD, Burns JM, Swerdlow RH. Feasibility and monoglycerides: individual effects and synergistic relationships - PubMed.
efficacy data from a ketogenic diet intervention in Alzheimer’s disease. Alz- Accessed, [Link] [Accessed 28 January
heimer’s Dement Transl Res Clin Interv. 2018;4:28e36. [Link] 2021].
10.1016/[Link].2017.11.002. [40] Novel antibacterial activity of monolaurin compared with conventional anti-
[15] Liang Z, Yue G, Ning X, et al. High dietary n-6/n-3 PUFA ratio promotes HDL biotics against organisms from skin infections: an in vitro study - PubMed.
cholesterol level, but does not suppress atherogenesis in apolipoprotein E- Accessed, [Link] [Accessed 28 January
null mice 1. J Atherosclerosis Thromb 2009;16(4):463e71. [Link] 2021].
10.5551/jat.no1347.

555