Alzheimer’s & Dementia: Translational Research & Clinical Interventions 4 (2018) 28-36

Featured Article

Feasibility and efficacy data from a ketogenic diet intervention in

Alzheimer’s disease
Matthew K. Taylora,b, Debra K. Sullivana,b, Jonathan D. Mahnkena,c, Jeffrey M. Burnsa,d,e,
Russell H. Swerdlowa,d,e,f,*
a
University of Kansas Alzheimer’s Disease Center, Fairway, KS, USA
b
Department of Dietetics and Nutrition, University of Kansas Medical Center, Kansas City, KS, USA
c
Department of Biostatistics, University of Kansas Medical Center, Kansas City, KS, USA
d
Department of Neurology, University of Kansas Medical Center, Kansas City, KS, USA
e
Department of Molecular and Integrative Physiology, University of Kansas Medical Center, Kansas City, KS, USA
f
Department of Biochemistry and Molecular Biology, University of Kansas Medical Center, Kansas City, KS, USA

Abstract Introduction: We assessed the feasibility and cognitive effects of a ketogenic diet (KD) in partici-
pants with Alzheimer’s disease.
Methods: The Ketogenic Diet Retention and Feasibility Trial featured a 3-month, medium-chain tri-
glyceride–supplemented KD followed by a 1-month washout in clinical dementia rating (CDR) 0.5,
1, and 2 participants. We obtained urine acetoacetate, serum b-hydroxybutyrate, food record, and
safety data. We administered the Alzheimer’s Disease Assessment Scale-cognitive subscale and
Mini–Mental State Examination before the KD, and following the intervention and washout.
Results: We enrolled seven CDR 0.5, four CDR 1, and four CDR 2 participants. One CDR 0.5 and all
CDR 2 participants withdrew citing caregiver burden. The 10 completers achieved ketosis. Most
adverse events were medium-chain triglyceride–related. Among the completers, the mean of the Alz-
heimer’s Disease Assessment Scale-cognitive subscale score improved by 4.1 points during the diet
(P 5 .02) and reverted to baseline after the washout.
Discussion: This pilot trial justifies KD studies in mild Alzheimer’s disease.
Ó 2017 The Authors. Published by Elsevier Inc. on behalf of the Alzheimer’s Association. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/
4.0/).
Keywords: Alzheimer’s disease; Cognition; Ketogenic diet; Low-carbohydrate diet; Medium-chain triglyceride

1. Introduction deoxyglucose positron emission tomography evaluations. Flu-

orodeoxyglucose positron emission tomography reveals focal
Aberrant energy metabolism occurs in Alzheimer’s disease
reductions in brain glucose utilization with advancing age
(AD) animal models and patients [1]. It manifests as perturbed
[2,3], which extend in magnitude and distribution during
mitochondrial function and structure as well as during fluoro-
AD [4,5]. Neuron loss, synaptic degradation, or bona fide
bioenergetic changes could contribute to this, although
M.K.T. and D.K.S. report no disclosures. J.D.M. is supported by
P30AG035982 and otherwise reports no disclosures. J.M.B. is supported
changes in asymptomatic subjects with AD risk factors
by P30AG035982 and receives or has received research support in the last suggest that bona fide changes exist [6–8]. Brain
2 years for clinical trials from Lilly, Avid Radiopharmaceuticals, Toyama homogenates of AD subjects also show less glucose
Chemical Company, Merck, and Biogen. R.H.S. is supported by consumption than control homogenates [9], even though ho-
P30AG035982 and within the last 2 years has received an honorarium mogenization disrupts synapses and neutralizes their impact;
from Accera and clinical trial support from Ausio Pharmaceuticals, LLC
and the Alzheimer’s Association.
persistence of reduced glucose flux in homogenates suggests
*Corresponding author. Tel.: 1-913-588-0555; Fax: 1-913-588-0681. that diminished synaptic connectivity does not exclusively ac-
E-mail address: rswerdlow@kumc.edu count for reduced AD glucose utilization.
https://doi.org/10.1016/j.trci.2017.11.002
2352-8737/ Ó 2017 The Authors. Published by Elsevier Inc. on behalf of the Alzheimer’s Association. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 M.K. Taylor et al. / Alzheimer’s & Dementia: Translational Research & Clinical Interventions 4 (2018) 28-36 29

In AD, targeting bioenergetic defects and decreased 2.2. Participants

glucose utilization for therapeutic purposes is reasonable
[10]. One potential approach includes the ketogenic diet The participants met McKhann et al. criteria for AD [23].
(KD). KDs increase fat and reduce carbohydrate con- Individuals were eligible to participate in the study if they
sumption [11]. This reduces insulin, which stimulates had a clinical dementia rating (CDR) of very mild AD
liver oxidation of fatty acids to ketone bodies (b-hydrox- (CDR 0.5), mild AD (CDR 1), or moderate AD (CDR 2).
ybutyrate [BHB], acetoacetate, and acetone) that enter the No clinically significant electrolyte, renal, or liver function
blood. BHB and acetoacetate access the brain and fuel abnormalities, a body mass index (BMI)
21 kg/m2, En-
respiration [12]. Acetone also circulates, it may access glish mastery, and an active study partner were also required.
the brain, and ventilation facilitates its excretion. Given Exclusion criteria included serious medical risks including
a choice between ketone bodies and glucose, neurons type 1 diabetes, ongoing or recent cancer, a cardiac event
preferentially consume ketone bodies [13–15]. Although in the past year, or other conditions deemed serious risks
brain glucose utilization declines in AD, ketone body by physicians on the study team. The KU Medical Center’s
utilization does not [16]. Institutional Review Board approved the protocol. Informed
Here, we considered the previously proposed question consent was obtained from all study participants as per insti-
[17] of whether a KD can benefit cognition in AD. tutional guidelines.
Although studies report that ketosis-like diets improve
memory in cognitively intact or mild cognitive impairment 2.3. The VHF-KD intervention and dietary intake
participants [18–20] and that a simple medium chain evaluations
triglyceride (MCT)-induced ketosis may improve AD Participants received nutrition counseling from the study
cognition [21,22], whether a true KD affects cognition in dietitian at the baseline study visit. Targeted macronutrient
actual AD patients or is even feasible remains unclear. composition for the dietary intervention included approxi-
Adding to the relevance of this question, KDs differ from mately 70% of energy as fat (including the MCT), 20% of en-
MCT-limited approaches in terms of ketone body pharma- ergy as protein, and restriction of carbohydrate to less than
cokinetics, dietary fat intake, insulin signaling effects, and 10% of energy; we sought a ketogenic ratio of 1:1 (ratio of
inflammation effects. We therefore tested the feasibility of lipid consumption in grams to nonlipid consumption in
maintaining AD participants on a KD and acquired prelim- grams) or better. Energy intake requirement and targeted
inary insight into how a KD affects cognition in AD daily MCT dosage was estimated using the Mifflin-St. Jeor
participants. equation [24]. A monthly supply of MCT oil (Now Foods,
USA), which contained a combination of C8:0 and C10:0
fatty acids, was provided at each study visit. In general, the
2. Methods
MCT dosage provided approximately 10% of total energy
2.1. Overall study design and recruitment from fat during the first week and increased by increments
of 10% during each successive week until reaching 40%,
The Ketogenic Diet Retention and Feasibility Trial while allowing for titration adjustments based on participant
(KDRAFT) was a single-arm, pilot clinical study with a tolerance. Participants could choose to consume the MCT
target enrollment of 15 participants with AD. The protocol neat or mix it with food or beverages. We provided daily
required participants to maintain an MCT-supplemented multivitamin, vitamin D, calcium, and phosphorus supple-
KD for 3 months and immediately on completing this inter- ments to help prevent micronutrient deficiencies. On comple-
vention to discontinue that diet and resume a normal diet for tion of the VHF-KD, participants were instructed to return to
1 month (thereby defining a 1-month washout period). a normal diet to complete a 1-month washout period.
Because from a calorie intake perspective our diet increased
dietary fat consumption and added MCT to offset lowered
2.4. Objectives and outcomes
carbohydrate consumption, to distinguish our diet from a
low-carbohydrate, calorie-restricted KD we herein refer to Our primary objective was to address the feasibility of
our diet intervention as a very high-fat KD (VHF-KD). implementing a VHF-KD intervention in AD participants.
To recruit for this trial, we first identified potential partic- Secondary objectives included evaluating the effects of a
ipants from a registry database maintained by the University VHF-KD on cognition.
of Kansas Alzheimer’s Disease Center. Then, we informed
potential participants about the study opportunity through 2.4.1. Dietary assessment
one of two approaches. One approach used letters or phone At the baseline visit, the study dietitian provided instruc-
calls to initiate discussion of this opportunity. Alternatively, tions to the participant’s study partner for completing 3-day
we invited potential participants to a KD cooking demon- food records (3DFRs). Each 3DFR recorded intake for two
stration at the University of Kansas (KU) Clinical Research weekdays and one weekend day. We entered 3DFR data
Center Demonstration Kitchen and used this event to initiate into the Nutrition Data System for Research (version
discussion of the research opportunity. 2016) to analyze nutrient intake. We collected 3DFRs
30 M.K. Taylor et al. / Alzheimer’s & Dementia: Translational Research & Clinical Interventions 4 (2018) 28-36

immediately after the baseline visit (just before starting the to quantify and characterize feasibility of the intervention.
diet); at months 1, 2, and 3; and during washout visits to Normality of the dependent variables was assessed through
characterize dietary intake before, during, and after the die- Q-Q visualization of residual models. Paired t-tests were
tary intervention. used to analyze differences in dietary macronutrient intake
before and during the diet intervention. We constructed
2.4.2. Biomarker, safety, and anthropometric assessments linear mixed models and tested mean differences with
Participants self-monitored urine ketones daily, in the post-hoc pairwise comparisons for cognitive data and serum
early evening, using urine acetoacetate test strips (Ketostix, biomarkers, with the exception of serum BHB where resid-
Bayer, Germany). Daily urine ketone status was recorded as ual assessment indicated the use of Friedman tests with Con-
either negative, trace (5–14.9 mg/dL), small (15–39.9 over post-hoc pairwise comparisons. Statistical analyses
mg/dL), moderate (40–79.9 mg/dL), or large (801 mg/dL) were performed using R (v. 3.3.2; R Foundation, Vienna,
in a provided diary. Days in which participants did not mea- Austria). Statistical tests were two-tailed, and significance
sure the ketone levels were conservatively tallied as a “nega- was set at P , .05. For the cognition analysis, our null hy-
tive” ketone response. pothesis assumed that a VHF-KD would not affect the cogni-
All serum biomarker and safety lab tests were collected af- tive performance of participants with AD.
ter a 12-hour fast. Full lipid and metabolic panels were
collected at the baseline and month 3 (end of diet intervention)
3. Results
visits, and these assays were performed by the KU Hospital
clinical laboratory. Serum electrolytes, renal function tests, Fifteen participants enrolled in the study over a 31-month
liver function tests, and glucose levels were measured by period (December 2013–July 2016). The mean age and educa-
the KU Hospital clinical laboratory at the baseline and month tion levels were 73.1 6 9.0 and 15.0 6 2.7 years, respectively.
3 visits. Serum BHB and insulin levels were measured at all Seven were male. To enroll these 15 participants, we contacted
visit time points (baseline, month 1, month 2, month 3, and 298 individuals. During the first 19 months, 72 potential partic-
washout) by the KU Hospital clinical laboratory. Homeostatic ipants were contacted by the study coordinator via letter or
model assessment 2-insulin resistance (HOMA2-IR) values phone call, and nine enrolled in the study. From months 20
for each participant were calculated using a HOMA2 Calcu- through 31, we invited 185 participants, by letter, to one of
lator (v. 2.2.3; University of Oxford, United Kingdom). two different ketogenic cooking demonstrations at the KU Clin-
At each visit, we queried the participants and their care part- ical Research Center Demonstration Kitchen. Thirty-eight total
ners for adverse symptoms. Also at each visit, we further potential participants attended one of two cooking demonstra-
encouraged the participants and their care partners to contact tions; attendance was limited by space limitations and not by
the study dietitian for any symptoms or medical complaints invitee interest. These two cooking demonstrations resulted
they thought might relate to their participation in the study. in the enrollment of six new participants. Supplementary
As part of an additional safety measure, participants received Figure 1 depicts recruitment and participant flow.
baseline and month 1, 2, and 3 electrocardiogram assessments. Five participants did not successfully implement the VHF-
Height, weight, and body composition were measured for KD, withdrew from the study within the first month or shortly
all subjects. BMI (kg/m2) was calculated using weight and after the first month (dropout rate of 33%), and provided only
height measurements. Dual energy x-ray absorptiometry baseline data. Four of the dropouts had moderate AD (CDR 2),
was used to attain and quantify body fat percentage, lean and the other had very mild AD (CDR 0.5) (Table 1). The part-
body mass, fat mass, and bone mass at baseline and month 3. ners of the participants that withdrew from the study uni-
formly cited caregiver burden as the reason for
2.4.3. Cognitive testing discontinuing. The remaining 10 participants were compliant
A trained psychometrician administered the Mini–Mental with the VHF-KD intervention, although one participant dis-
State Examination (MMSE) and the Alzheimer’s Disease continued cholinesterase inhibitor therapy and was therefore
Assessment Scale-cognitive subscale (ADAS-cog) at baseline, technically protocol noncompliant. This was the only partic-
at the end of the intervention (month 3), and after the 1-month ipant in whom we documented cognitive decline.
washout. The MMSE is a brief, 30-point questionnaire de- Most participants preferred to mix their MCT oil with
signed to identify cognitive impairment. The ADAS-cog con- either food or beverages and used this strategy to enhance
sists of 11 tasks that measure changes in memory, language, MCT tolerability. Participants who completed the interven-
praxis, and attention. Total incorrect answers and cued re- tion generally consumed 1.5–3 tablespoons of MCT oil per
minders are reported as the cumulative score, thus higher day, which constituted 60–80% of the target MCT volume.
scores on the ADAS-cog reflect greater cognitive impairment. No one reached 100% of the MCT target volume.
Self-reported urine ketone assessments from the 10 com-
pleters revealed each achieved ketosis, although the depth
2.5. Statistical analyses
and consistency of ketosis varied. On average, completers
Continuous variables were described using their means detected the presence of at least some degree of urine acetoa-
and standard deviations. Descriptive statistics were utilized cetate on 54.5 6 29.0 days (60.6%) of the 3-month dietary
 M.K. Taylor et al. / Alzheimer’s & Dementia: Translational Research & Clinical Interventions 4 (2018) 28-36 31

Table 1 cholesterol, blood glucose, insulin, and homeostatic model

Participants and trial disposition assessment 2-insulin resistance parameters remained stable.
Subject Data for Time/reason The VHF-KD did not alter electrolyte, renal, or liver func-
ID CDR Sex analysis for dropout Symptoms/complaints tion values. BMI, body fat percentage, and bone mass did
Very Mild AD not change. There were no electrocardiogram changes.
01* 0.5 M X Diarrhea We used a linear mixed model with pairwise analysis to
02 0.5 M Week 2/Caregiver compare cognitive performance at baseline, month 3, and after
Burden
05y 0.5 M X
the 1-month washout period (Fig. 2). Including all diet
06 0.5 F X Nausea, Diarrhea compliant participants (n 5 10), ADAS-cog scores signifi-
07 0.5 F X cantly changed from baseline to month 3 with a mean improve-
11 0.5 F X ment of 4.1 points (25.5 vs. 21.4, P 5 .02). Excluding the one
13 0.5 M X Diarrhea protocol noncompliant completer (n 5 9), ADAS-cog scores
Mild AD
08 1 F X
changed from baseline to month 3 with a mean improvement
10 1 M X Diarrhea of 5.3 points (26.6 vs. 21.3, P 5 .001). ADAS-cog scores re-
14 1 F X Diarrhea verted to baseline following the 1-month washout period. After
15 1 F X the washout period, the mean ADAS-cog score was 25.3 for all
Moderate AD participants (P 5 .09 for the comparison between the 3-month
03 2 M Week 4/Caregiver
Burden
and the washout values) and also for the protocol compliant
04 2 F Week 4/Caregiver subjects (P 5 .05 for the comparison between the 3 month
Burden and washout values). Excluding the protocol noncompliant
09 2 M Week 5/Caregiver participant, MMSE scores significantly improved from base-
Burden line to month 3 (25.2 vs. 26.3, P 5 .05) (Table 3).
12 2 F Week 4/Caregiver
Burden
Abbreviations: CDR, clinical dementia rating; AD, Alzheimer’s disease; 4. Discussion
M, male; F, female.
*Subject 01 completed the intervention, but did not contribute washout
This pilot trial investigated the feasibility of a 3-month
cognitive data. MCT-supplemented KD intervention in AD participants.
y
Subject 05 was intervention-compliant, but not protocol compliant as The VHF-KD proved feasible in very mild (CDR 0.5) and
cholinesterase inhibitor therapy was discontinued. mild (CDR 1) participants, as 10 of 11 adapted to a restricted
carbohydrate and augmented fat nutrient pattern, reached the
intervention. Forty-nine percent of the positive daily read- target macronutrient ratio, demonstrated elevated urine and
ings were reported as trace, 26% as small, 23% as moderate, serum ketone levels, and completed the intervention. We
and 2% as large. Fig. 1A summarizes each subject’s urine cannot generalize this conclusion to moderate (CDR 2) par-
ketone data. ticipants, though, as all four CDR 2 participants failed to
Serum BHB levels were significantly elevated at months 1, achieve ketosis and each withdrew from the study. Second-
2, and 3 compared with that of baseline (0.11 mmol/L at base- arily, study completers demonstrated objective improvement
line vs. 0.52, 0.34, and 0.31 mmol/L; P , .001) (Fig. 1B and on a cognitive assessment (the ADAS-cog), and this objec-
1C) and returned to the normal range at the end of the washout tive improvement disappeared after a washout period.
period (0.12 mmol/L). Serum BHB results generally seemed Existing literature documents the challenges of enrolling
to reflect urine ketone responses with one exception. One par- adults into KD studies [25,26]. Perceptions of the KD border
ticipant’s (subject 15) urine acetoacetate reached only trace on undesirable, which make it difficult to recruit participants
levels and for just 24.2% of the days on the diet, but via traditional contact methods (i.e. letter, phone call, and
throughout the intervention, serum BHB measurements clinic visit). Placing persons with AD on a KD presents an
showed a relatively robust induction (see Fig. 1A and 1B). additional hurdle, as this inevitably contributes to care-
Table 2 shows values for anthropometric measures, serum provider burden. Based on our study’s recruitment experience,
biomarkers, safety labs, metabolic measures, and dietary we believe that KD studies in adult populations require unique,
intake. Preintervention and intraintervention energy intake interactive strategies to generate participation interest. We
values were comparable. During the intervention, total fat found that KD cooking demonstrations enhanced interest,
intake increased (fat: 90.6 g vs. 166.7 g; P , .001), and total altered KD preconceptions, and expedited recruitment.
carbohydrate intake was significantly reduced (209.8 g vs. Participants with more advanced dementia did not comply
46.0 g; P , .001). The mean macronutrient proportion dur- with the diet intervention. A priori, one might predict that a
ing the VHF-KD intervention was 73.4% fat, 9% carbohy- positive correlation could exist between dementia severity
drate, and 17.6% protein as energy. and compliance because dementia severity should also corre-
No serious adverse events occurred. Participant com- late inversely with the diversity of food choice requests. In
plaints were minor in magnitude and transient, and mostly other words, moderately or severely demented individuals
limited to MCT-associated diarrhea (in 50%). Weight, might generate fewer requests for specific meals or snacks
32 M.K. Taylor et al. / Alzheimer’s & Dementia: Translational Research & Clinical Interventions 4 (2018) 28-36

Fig. 1. Urine and serum BHB values. (A) The figure indicates the percentage of days in which each participant’s self-reported urine acetoacetate analysis re-
vealed ketosis, as well as the depth of ketosis achieved. (B) Serum BHB values for each subject. Subject 15, who reported only trace urine acetoacetate for only
24.2% of the intervention, is shown by a yellow line. (C) Mean serum BHB values 6 SD. ***P , .001. Abbreviations: BHB, b-hydroxybutyrate; SD, standard
deviation; ADAS-cog, Alzheimer’s Disease Assessment Scale-cognitive subscale; CDR, clinical dementia rating.
 M.K. Taylor et al. / Alzheimer’s & Dementia: Translational Research & Clinical Interventions 4 (2018) 28-36 33

Table 2 comply with the intervention. The relatively small size of

Anthropometric, blood-derived, and dietary intake data (n 5 10) this study may account for our inability to identify specific
Parameter Baseline Month 3 P-value characteristics that predict compliance.
Anthropometric measures A higher rate of study withdrawal among those with mod-
Weight, kg 76.5 (17.4) 74.7 (16.2) .81 erate dementia suggests that KD studies of community-
BMI, kg/m2 27.7 (7.5) 26.9 (6.8) .81 based AD participants should focus on individuals with
Body fat, % 38.6 (11.5) 38.2 (11.3) .94 mild and very mild dementia. Studies of moderate or severe
Lean body mass, kg 43.5 (7.6) 42.8 (8.6) .85
Bone mass, kg 2.7 (0.7) 2.7 (0.7) .98
AD may still be possible, but should perhaps consider
Lipid status limiting the intervention to participants that live in institu-
Total cholesterol, mg/dL 173.0 (27.5) 194.4 (36.3) .16 tional settings (to minimize diet-associated increases in
LDL, mg/dL 102.1 (18.7) 117.4 (25.4) .14 caregiver burden).
HDL, mg/dL 55.4 (9.2) 58.2 (8.6) .49 Previously reported KD side effects include constipation,
Serum biomarkers
AST, U/L 21.1 (5.2) 24.5 (7.0) .24
nausea, vomiting, diarrhea, fatigue, and hunger [28]. No
ALT, U/L 15.7 (6.5) 18.9 (8.1) .34 participant complained of hunger or fatigue. Some of our
Alkaline phosphatase, U/L 62.6 (21.3) 55.8 (19.5) .48 participants did report gastrointestinal side effects, particu-
Calcium, mg/dL 9.4 (0.4) 9.5 (0.3) .71 larly diarrhea. Participants and their care-partners, though,
Chloride, mEq/L 104.1 (1.3) 105.4 (2.2) .12 tended to associate these complaints with MCT oil consump-
Potassium, mEq/L 4.0 (0.3) 4.1 (0.2) .47
Sodium, mEq/L 137.2 (1.8) 137.9 (1.9) .41
tion and not the KD itself. Utilization of a flexible MCT titra-
BUN, mg/dL 20.9 (4.3) 20.1 (6.1) .74 tion protocol, therefore, likely enhanced compliance and
Creatinine, mg/dL 1.0 (0.2) 1.1 (0.3) .58 reduced adverse event quality and quantity.
Metabolic status There were no serious adverse events. Laboratory and elec-
Insulin, mU/mL 5.1 (2.2) 5.0 (2.2) .94 trocardiogram tests raised no safety concerns. Weights re-
Glucose, mg/dL 92.6 (5.6) 95.8 (10.0) .40
HOMA2-IR 0.7 (0.3) 0.7 (0.3) .97
mained stable and there was no loss of bone mass. We
Beta, % 67.8 (22.1) 62.3 (19.0) .56 conclude that a VHF-KD is safe in AD, at least over 3 months,
Sensitivity, % 169.6 (51.0) 182.3 (86.4) .70 and that persons with AD can maintain and tolerate a VHF-
Dietary intake KD for at least 3 months.
Energy (kcal) 1950.7 (450.6) 2002.8 (660.7) .80 Because of our small sample size and single-arm design,
Fat, g 90.6 (23.5) 166.7 (60.0) ,.001*
Carbohydrate, g 209.8 (80.3) 46.0 (26.9) ,.001*
any interpretation of this study’s cognitive performance data
Protein, g 81.3 (21.1) 90.0 (28.2) .34 requires caution. We may parsimoniously state, though, that
a statistically significant improvement in ADAS-cog score
Abbreviations: BMI, body mass index; LDL, low-density lipoprotein;
HDL, high-density lipoprotein; AST, aspartate aminotransferase; ALT,
coincident with the diet intervention was observed, which is
alanine aminotransferase; BUN, blood urea nitrogen; HOMA2-IR, homeo- consistent with our research hypothesis of diet-associated
static model assessment 2-insulin resistance; SD, standard deviation. changes in this measure. This apparent improvement could
NOTE. Values are Mean (SD). reflect a placebo effect, the expectations of an unblinded psy-
*Significant difference between time points (P  .05). chometrician, participant motivation, or a test–retest effect
although reversion of the ADAS-cog scores after a washout
than mildly demented individuals. Conversely, dementia period argue against the possibility of a test–retest effect.
severity also positively correlates with amounts of caregiver Only one participant’s ADAS-cog score declined while
burden [27]. Our KD intervention required study partners to following the diet protocol. This participant’s performance
assume varying degrees of additional responsibility, including continued to deteriorate for at least 1 month after completing
compliance with complex meal preparation instructions, MCT the KD intervention, so we do not believe that the KD itself
administration, urinary ketone assessment, and food diary was responsible. Moreover, this participant discontinued
completion. Implementing a relatively complex intervention cholinesterase inhibitor therapy shortly after starting the
such as a KD may have simply overwhelmed the caregivers KD and was therefore technically protocol noncompliant.
of the moderate AD participants. Our data are consistent We nevertheless included this participant’s data in this report
with this scenario. It is important to note, though, our impres- for completeness and transparency.
sion that high levels of study partner burden influenced partic- The KDRAFT complements other studies that report keto-
ipant withdrawal is based on informal, anecdotal study partner genic interventions benefit cognitive performance. A 3-month
reporting and not on formal measurements of study partner duration caloric restriction trial found that CR improved mem-
burden. Future studies of a KD in AD should consider ory in healthy adults [18]. Although the investigators did not
applying quantitative measurements of caregiver burden. assess ketone body production, caloric restriction does induce
When the study partner was a male, the dropout rate was ketosis [29]. Another study reported a single MCT-
25%, and when the study partner was a female, the dropout supplemented “ketogenic meal” acutely boosted cognitive per-
rate was 43%. These percentages are not significantly formance in nondemented elderly adults [20]. An exploratory
different. In general, we could not identify differences be- study randomized 23 subjects with mild cognitive impairment,
tween study partners of participants who did and did not a frequent AD-precursor state, to a 6-week carbohydrate-
34 M.K. Taylor et al. / Alzheimer’s & Dementia: Translational Research & Clinical Interventions 4 (2018) 28-36

Fig. 2. ADAS-cog scores. (A) Waterfall plot showing the difference between the baseline and 3-month VHF-KD ADAS-cog scores for the 10 participants who
completed the diet intervention. Subject 15 was CDR 1. (B) The ADAS-cog scores for each participant who completed the diet intervention (baseline score, after
3 months on the VHF-KD, and after a 1 month washout period) are shown. One of the CDR 0.5 participants did not complete washout testing and so an ADAS-
cog washout value is not available for this participant. (C) Mean change 6 SD ADAS-cog data for all participants who completed the VHF-KD intervention. (D)
Mean change 6 SD ADAS-cog data for the protocol-compliant participants. *P  .05; ***P , .005. Abbreviations: ADAS-cog, Alzheimer’s Disease Assess-
ment Scale-cognitive subscale; VHF-KD, very high-fat KD; CDR, clinical dementia rating; SD, standard deviation.

restricted or high-carbohydrate diet and found memory test KDs and ketone bodies have protean physiologic effects.
performance improved in the carbohydrate-restricted KDs can increase ketone body levels and ketone utilization,
(n 5 12) but not in the high-carbohydrate intervention [19]. reduce brain glucose consumption, lower insulin, alter insu-
Although the authors did not formally refer to their lin signaling, increase long- and medium-chain fatty acids,
carbohydrate-restricted diet as a KD, they did show that cogni- affect lipid handling, and reduce inflammation [11,31–37].
tive improvement positively correlated with ketone body Ketone bodies change bioenergetic infrastructures in
levels. Two studies report that simply consuming an MCT sup- neurons and astrocytes, mediate glia-neuron interactions,
plement outside of a KD context, which also boosts serum affect energy homeostasis, post-translationally modify pro-
BHB levels, at least temporarily improves cognition in AD teins (directly and indirectly) to influence their function,
subjects [21,22]. In a case report, Newport et al. noted modify gene expression, and act as signaling molecules
administering a BHB-promoting ketone monoester to an AD [31,38–42]. Our study does not address the mechanisms
patient coincided with cognitive improvement [30]. through which a KD might influence cognition. It does not

Table 3
KDRAFT cognitive test scores
All participants Protocol compliant participants
Parameter Baseline (n 5 10) Month 3 (n 5 10) Washout (n 5 9) Baseline (n 5 9) Month 3 (n 5 9) Washout (n 5 8)
y
ADAS-cog 25.5 (5.9) 21.4 (4.4)* 25.3 (5.4) 26.6 (5.1) 21.3 (4.7) 25.3 (5.7)
MMSE 25.5 (1.5) 26.3 (0.5) 25.1 (1.8) 25.2 (1.3) 26.3 (0.5)z 25.4 (1.7)
Abbreviations: KDRAFT, Ketogenic Diet Retention and Feasibility Trial; ADAS-cog, Alzheimer’s Disease Assessment Scale-cognitive subscale; MMSE,
Mini–Mental State Examination; KD, ketogenic diet; SD, standard deviation.
NOTE. Values are Mean (SD).
*P 5 .02, Baseline vs. KD by linear mixed model.
y
P 5 .001, Baseline vs. KD by linear mixed model.
z
P 5 .05, Baseline vs. KD by linear mixed model.
 M.K. Taylor et al. / Alzheimer’s & Dementia: Translational Research & Clinical Interventions 4 (2018) 28-36 35

address whether KDs and non-KD ketogenic interventions References

are functionally or therapeutically interchangeable.
Because of the KDRAFT’s exploratory nature, small size, [1] Swerdlow RH. Mitochondria and cell bioenergetics: increasingly
and single-arm structure, we cannot definitively conclude recognized components and a possible etiologic cause of Alzheimer’s
disease. Antioxid Redox Signal 2012;16:1434–55.
that a KD benefits AD patients. Feasibility and efficacy
[2] De Santi S, de Leon MJ, Convit A, Tarshish C, Rusinek H, Tsui WH,
data from this 3-month pilot trial, though, justify further et al. Age-related changes in brain: II. Positron emission tomography
studies of KDs in mild AD participants. Although positive of frontal and temporal lobe glucose metabolism in normal subjects.
results from definitive research studies might lead some to Psychiatr Q 1995;66:357–70.
consider a KD for the treatment of AD, in the routine clinical [3] Nugent S, Tremblay S, Chen KW, Ayutyanont N, Roontiva A,
Castellano CA, et al. Brain glucose and acetoacetate metabolism: a
setting, its challenges may hinder its implementation. On a
comparison of young and older adults. Neurobiol Aging 2014;
fundamental level, definitive positive studies would never- 35:1386–95.
theless provide insight into and justify evaluating other [4] Ferris SH, de Leon MJ, Wolf AP, Farkas T, Christman DR, Reisberg B,
metabolic manipulations for the treatment of AD. et al. Positron emission tomography in the study of aging and senile
dementia. Neurobiol Aging 1980;1:127–31.
[5] Foster NL, Chase TN, Fedio P, Patronas NJ, Brooks RA, Di Chiro G.
Acknowledgments Alzheimer’s disease: focal cortical changes shown by positron emis-
sion tomography. Neurology 1983;33:961–5.
This study was supported by the University of Kansas Alz-
[6] Reiman EM, Caselli RJ, Yun LS, Chen K, Bandy D, Minoshima S,
heimer’s Disease Center (P30AG035982), the University of et al. Preclinical evidence of Alzheimer’s disease in persons homozy-
Kansas Heartland Clinical Translational Science Award Cen- gous for the epsilon 4 allele for apolipoprotein E. N Engl J Med 1996;
ter, the University of Kansas Department of Dietetics and 334:752–8.
Nutrition, and the Stop Alzheimer’s Now Foundation. [7] Small GW, Mazziotta JC, Collins MT, Baxter LR, Phelps ME,
Mandelkern MA, et al. Apolipoprotein E type 4 allele and cerebral
glucose metabolism in relatives at risk for familial Alzheimer disease.
Supplementary data JAMA 1995;273:942–7.
[8] Mosconi L, Brys M, Switalski R, Mistur R, Glodzik L, Pirraglia E,
Supplementary data related to this article can be found at et al. Maternal family history of Alzheimer’s disease predisposes to
https://doi.org/10.1016/j.trci.2017.11.002. reduced brain glucose metabolism. Proc Natl Acad Sci U S A 2007;
104:19067–72.
[9] Swerdlow R, Marcus DL, Landman J, Kooby D, Frey W 2nd,
Freedman ML. Brain glucose metabolism in Alzheimer’s disease.
RESEARCH IN CONTEXT Am J Med Sci 1994;308:141–4.
[10] Swerdlow RH. Bioenergetic medicine. Br J Pharmacol 2014;
171:1854–69.
[11] Kossoff EH, Zupec-Kania BA, Rho JM. Ketogenic diets: an update for
1. Systematic review: Literature was reviewed using child neurologists. J Child Neurol 2009;24:979–88.
PubMed and revealed, although proposed as a poten- [12] Owen OE, Morgan AP, Kemp HG, Sullivan JM, Herrera MG,
tial therapy, no published studies assessing feasibility Cahill GF Jr. Brain metabolism during fasting. J Clin Invest 1967;
46:1589–95.
and efficacy of a ketogenic diet therapy in humans
[13] Edmond J, Robbins RA, Bergstrom JD, Cole RA, de Vellis J. Capacity
with Alzheimer’s disease (AD). for substrate utilization in oxidative metabolism by neurons, astro-
cytes, and oligodendrocytes from developing brain in primary culture.
2. Interpretation: We investigated the feasibility of a J Neurosci Res 1987;18:551–61.
medium-chain triglyceride–supplemented KD inter- [14] Hasselbalch SG, Knudsen GM, Jakobsen J, Hageman LP, Holm S,
vention in AD participants and coincident effects on Paulson OB. Blood-brain barrier permeability of glucose and ketone
cognition. The dietary protocol was feasible in very bodies during short-term starvation in humans. Am J Physiol 1995;
mild (clinical dementia rating [CDR] 0.5) and mild 268:E1161–6.
[15] Cunnane SC, Courchesne-Loyer A, Vandenberghe C, St-Pierre V,
(CDR 1) AD patients. All patients with moderate Fortier M, Hennebelle M, et al. Can ketones help rescue brain fuel sup-
(CDR 2) AD withdrew from the study, citing caregiver ply in later life? Implications for cognitive health during aging and the
burden. Alzheimer’s Disease Assessment Scale- treatment of Alzheimer’s disease. Front Mol Neurosci 2016;9:53.
cognitive subscale scores significantly improved in [16] Castellano CA, Nugent S, Paquet N, Tremblay S, Bocti C, Lacombe G,
dietary compliant participants from baseline to the end et al. Lower brain 18F-fluorodeoxyglucose uptake but normal 11C-
acetoacetate metabolism in mild Alzheimer’s disease dementia. J Alz-
of the intervention and returned to baseline level after
heimers Dis 2015;43:1343–53.
a 1-month discontinuance of the diet. Our findings [17] Swerdlow R, Marcus DM, Landman J, Harooni M, Freedman ML.
suggest that the medium chain triglyceride- Brain glucose and ketone body metabolism in patients with Alz-
supplemented KD is feasible in AD patients up to heimer’s disease. Clin Res 1989;37:461A.
CDR 1 and may elicit potential cognitive benefit. [18] Witte AV, Fobker M, Gellner R, Knecht S, Floel A. Caloric restriction
improves memory in elderly humans. Proc Natl Acad Sci U S A 2009;
3. Future directions: These findings justify further inves- 106:1255–60.
tigation into the ketogenic diet as a therapy in AD. [19] Krikorian R, Shidler MD, Dangelo K, Couch SC, Benoit SC, Clegg DJ.
Dietary ketosis enhances memory in mild cognitive impairment. Neu-
robiol Aging 2012;33:425.e19–27.
36 M.K. Taylor et al. / Alzheimer’s & Dementia: Translational Research & Clinical Interventions 4 (2018) 28-36

[20] Ota M, Matsuo J, Ishida I, Hattori K, Teraishi T, Tonouchi H, et al. Ef- [30] Newport MT, VanItallie TB, Kashiwaya Y, King MT, Veech RL. A new
fect of a ketogenic meal on cognitive function in elderly adults: poten- way to produce hyperketonemia: use of ketone ester in a case of Alz-
tial for cognitive enhancement. Psychopharmacology (Berl) 2016; heimer’s disease. Alzheimers Dement 2015;11:99–103.
233:3797–802. [31] Puchalska P, Crawford PA. Multi-dimensional roles of ketone bodies
[21] Reger MA, Henderson ST, Hale C, Cholerton B, Baker LD, in fuel metabolism, signaling, and therapeutics. Cell Metab 2017;
Watson GS, et al. Effects of beta-hydroxybutyrate on cognition in 25:262–84.
memory-impaired adults. Neurobiol Aging 2004;25:311–4. [32] Selfridge JE, Wilkins HM, Lezi E, Carl SM, Koppel S, Funk E, et al.
[22] Henderson ST, Vogel JL, Barr LJ, Garvin F, Jones JJ, Costantini LC. Effect of one month duration ketogenic and non-ketogenic high fat di-
Study of the ketogenic agent AC-1202 in mild to moderate ets on mouse brain bioenergetic infrastructure. J Bioenerg Biomembr
Alzheimer’s disease: a randomized, double-blind, placebo-controlled, 2015;47:1–11.
multicenter trial. Nutr Metab (Lond) 2009;6:31. [33] Dupuis N, Curatolo N, Benoist JF, Auvin S. Ketogenic diet exhibits
[23] McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR Jr, anti-inflammatory properties. Epilepsia 2015;56:e95–8.
Kawas CH, et al. The diagnosis of dementia due to Alzheimer’s dis- [34] McDaniel SS, Rensing NR, Thio LL, Yamada KA, Wong M. The keto-
ease: Recommendations from the National Institute on Aging-Alz- genic diet inhibits the mammalian target of rapamycin (mTOR)
heimer’s Association workgroups on diagnostic guidelines for pathway. Epilepsia 2011;52:e7–11.
Alzheimer’s disease. Alzheimers Dement 2011;7:263–9. [35] Bough KJ, Wetherington J, Hassel B, Pare JF, Gawryluk JW,
[24] Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. A Greene JG, et al. Mitochondrial biogenesis in the anticonvulsant mech-
new predictive equation for resting energy expenditure in healthy in- anism of the ketogenic diet. Ann Neurol 2006;60:223–35.
dividuals. Am J Clin Nutr 1990;51:241–7. [36] Hartman AL, Gasior M, Vining EP, Rogawski MA. The neuropharma-
[25] Cervenka MC, Henry BJ, Felton EA, Patton K, Kossoff EH. Establish- cology of the ketogenic diet. Pediatr Neurol 2007;36:281–92.
ing an Adult Epilepsy Diet Center: Experience, efficacy and chal- [37] Noh HS, Lee HP, Kim DW, Kang SS, Cho GJ, Rho JM, et al. A cDNA
lenges. Epilepsy Behav 2016;58:61–8. microarray analysis of gene expression profiles in rat hippocampus
[26] Mosek A, Natour H, Neufeld MY, Shiff Y, Vaisman N. Ketogenic diet following a ketogenic diet. Brain Res Mol Brain Res 2004;129:80–7.
treatment in adults with refractory epilepsy: a prospective pilot study. [38] Guzman M, Blazquez C. Is there an astrocyte-neuron ketone body
Seizure 2009;18:30–3. shuttle? Trends Endocrinol Metab 2001;12:169–73.
[27] Yu H, Wang X, He R, Liang R, Zhou L. Measuring the caregiver [39] Newman JC, Verdin E. b-hydroxybutyrate: much more than a metab-
burden of caring for community-residing people with Alzheimer’s dis- olite. Diabetes Res Clin Pract 2014;106:173–81.
ease. PLoS One 2015;10:e0132168. [40] Menzies KJ, Zhang H, Katsyuba E, Auwerx J. Protein acetylation in meta-
[28] Neal EG, Chaffe H, Schwartz RH, Lawson MS, Edwards N, bolism - metabolites and cofactors. Nat Rev Endocrinol 2016;12:43–60.
Fitzsimmons G, et al. A randomized trial of classical and medium- [41] Xie Z, Zhang D, Chung D, Tang Z, Huang H, Dai L, et al. Metabolic
chain triglyceride ketogenic diets in the treatment of childhood epi- regulation of gene expression by histone lysine beta-hydroxybutyryla-
lepsy. Epilepsia 2009;50:1109–17. tion. Mol Cell 2016;62:194–206.
[29] Lin AL, Zhang W, Gao X, Watts L. Caloric restriction increases ketone [42] Cullingford TE. The ketogenic diet; fatty acids, fatty acid-activated re-
bodies metabolism and preserves blood flow in aging brain. Neurobiol ceptors and neurological disorders. Prostaglandins Leukot Essent
Aging 2015;36:2296–303. Fatty Acids 2004;70:253–64.