TABLE 2.

Studies analyzing the safety and efficacy of KD with evidence of sustained ketosis.

Study	Study type	Disease	Intervention	Results	Measuring ketosis	Safety of KD
Animal models
Streijger et al., 2013	C5 hemi-section rodent model	Spinal cord injury	KD 3:1 4 h following injury	The study associated KD to smaller spinal lesion sizes and improved ipsilateral forelimb movement, and, interestingly, when MCT was pharmacologically inhibited, the neuroprotective effect by KD was prevented	Blood Ketone concentrations (mmol/L), specially βHB, via Medisense Precision Xtra monitor. Rodents sustained ketosis for 12 weeks after injury.	Not addressed
Kong et al., 2017	C5-hemisection rodent model	Spinal cord injury	KD 7:1 2 weeks prior to injury	KD metabolite beta-Hydroxybutyrate administration can mitigate oxidative stress in spinal cord injury through KD’s suppression of class I histone deacetylases.	Measured blood ketone levels (βHB) – Two weeks prior to injury, ketone levels reached ∼2.8 mM, then decreased to ∼1.5 mM the day after SCI, and then slightly increased at 3 days after SCI. At 7 days post-injury, ketone level increased to pre-injury levels (∼2.8 mM).	Not addressed
Tai et al., 2008	Rodent model	Cardiac arrest induced cerebral ischemia	KD fat (78%), protein (10%), carbohydrate (2%), and inert components (10%).	KD prevented cardiac arrest-induced cerebral ischemic events, as well as neurodegeneration in the cerebellum and thalamic reticular nucleus.	Sustained ketosis for 25 days in KD fed rats was confirmed by colorimetric determination of the blood βHB level using a βHB LiquiColor kit	Not addressed
Shaafi et al., 2019	Stereotactic endothelin-1 (ET-1) MCA stroke model	Ischemic stroke	KD 4:1 + 3.5 cc/day MCT oil + 12 h fast before to induce ketosis quicker, given 3 days prior to stroke induction	Preconditioning KD significantly reduced motor dysfunction as measured by all motor-behavior function assessments.	KB levels measured prior to KD, and repeated on 2nd, 4th, 6th, 9th, 11th day of the study. KB levels were significantly higher than the other two groups from the second day and sustained consistently during the remainder of the study (3.39 ± 0.81 mmol/L, 0.31 ± 0.09 and 0.34 ± 0.07 mmol/L respectively, P < 0.001).	Not addressed
Ma et al., 2018	Young healthy mice	Neurovascular integrity	KD feed (not specified) over 16 weeks	At 16 weeks, KD fed mice had significant increases in cerebral blood flow, BBB P-glycoprotein channel transporter’s, reduced mTOR, and enhanced expression of endothelial nitric oxide syntheses. Changes in microbiota was noted – increases in Akkermansia muciniphila and Lactobacillus and decreases in Desulfovibrio and Turicibacter.	Measured blood ketone levels - KD fed mice had significantly higher (43%) ketone levels compared to control mice (p = 0.0004). Ketone and glucose levels were inversely related.	Not addressed
Ari et al., 2014	SOD1-G93A transgenic ALS mice	ALS	KD custom-diet with macronutrient profile similar to human prescribed KD. Started diet at 10 weeks old and continued until survival endpoint.	KD-fed mice had greater motor performance and extended survival time than controls fed a standard diet.	Blood collected from tail snip of glucose and βHB at survival endpoint. Blood βHB was over 50% higher in KD fed mice.	Not addressed
Zhao et al., 2006	SOD1-G93A transgenic ALS mice	ALS	KD feed based on known human formulation	KD has been associated with augmented motor unit survival and motor performance improvement.	KD fed mice showed > 3.5 elevation in the blood concentration of circulating ketone bodies (acetone, acetoacetate, and βHB) compared to standard feed animals (1056 ± 197 vs. 360 ± 43 μM, p = 0.012). The principle KB (78%) measured was is βHB.	Not addressed
Human studies
Phillips et al., 2018, 2021; Taylor et al., 2018; White et al., 2020	RCT, crossover trial	Alzheimer’s Disease	Low fat diet versus ketogenic diet recipes which constituted 58% fat, 29% protein, 7% fiber, and 6% net carbohydrate by weight - 12 weeks intervention with 10 weeks washout period	Modified KD achieved sustained ketosis during KD period and cognitive performance improvement in AD Cooperative Study - Activities of Daily Living (ADCS-ADL) inventory, and Quality of Life in AD (QOL-AD) questionnaire in those with Alzheimer’s disease.	Patients who completed the ketogenic diet reached sustained physiological ketosis (12-week mean beta-hydroxybutyrate level: 0.95 ± 0.34 mmol/L).	Mild adverse effects – irritability, fatigue, sugar cravings. Irritability is a common adverse effect of AD [28].
Nagpal et al., 2019	Double-blind, cross-over, pilot RCT	Alzheimer’s disease	MMKD versus American Heart Association Diet (AHAD) - 6-week diet intervention with 6 weeks washout period.	MMKD increased abundance of Enterobacteriaceae, Akkermansia, Slackia, Christensenellaceae and Erysipelotrichaceae and decreased levels of Bifidobacterium and Lachnobacterium. MMKD had minor decreases in fecal lactate and acetate but increases SCFAs propionate and butyrate.	Blood ketone levels measured weekly to ensure compliance – 17 of 23 initiating the diets were included in the fecal analysis	No adverse effects or safety of KD was addressed
Nagpal et al., 2020	Randomized, double-blind, crossover, single-center pilot trial	Mild cognitive impairment (MCI)	MMKD versus American Heart Association Diet (AHAD) - 6-week diet intervention with 6 weeks washout period.	MMKD induced a broader effect on gut fungal (mycobiome) diversity in subjects with MCI compared to cognitively normal controls. MMKD on MCI also increased Agaricus and Mrakia while decreased Saccharomyces and Claviceps fungal taxons.	Not addressed	Not addressed
Taylor et al., 2018	Pilot RCT	Alzheimer’s disease	MCT supplemented KD aka “Very high-fat KD” (VHFKD) – 70% fat including MCT, 1:1 ketogenic ratio, 8 weeks	Study confirmed ketosis and cognitive improvement over the 3-month period, noted cognitive reversion to baseline during the washout phase, and justified safety of KD in AD.	Urine acetoacetate levels were measured daily by participants and βHB plasma levels were measured monthly. Study noted significant increases in βHB levels during the VHFKD period but urine results did not suggest consistent acetoacetate levels throughout the diet (60.6% of participants resulted some degree of acetoacetate in their urine)	MCT-associated diarrhea.
Phillips et al., 2018	Pilot RCT	Parkinson’s disease	Low-fat diet vs. KD. KD meal plan consisted of daily intake of 1,750 kcal per day composed of 42 g of fat, 75 g of protein, 246 g net carbohydrate, and 33 g of fiber.	PD patients adhering to KD diet sustained ketosis (measured via βHB) and had significantly decreased MDS-Unified Parkinson’s Disease Rating Scale scores and greater improvements in non-motor symptoms (cognitive impairment, fatigue, pain sensations) compared to patients fed a low-fat diet.	Measured βHB levels. Over 8 weeks, results saw significant differences between the diet groups in weekly mean bedtime blood glucose (low−fat group: 6.28 ± 0.73 mmol/L vs. ketogenic group: 5.70 ± 1.20 mmol/L; P = 0.001) and ketone (0.16 ± 0.05 mmol/L vs. 1.15 ± 0.59 mmol/L; P < 0.001) levels.	Minor exacerbation of tremor and/or rigidity - 50% of patients week 1–4 then reduced to 29% in weeks 5–8.
White et al., 2020	Phase II RCT intervention.	Acute brain injury – phase II intervention - enrolled within 48 h of admission	KD combo enteral feed via NG tube: KetoCall® (4:1 Ratio, 88.7% fat, 8.3% protein, 3.1% CHO) + Protifar® (protein feed 28 g/l to meet caloric goal)	The study noted that ketone levels increased with KD enteral feed and that KB levels correlated with KetoCal caloric intake. The study concluded the safety of KD tube feeds on adult brain injury (no acid/base changes, hypoglycemia, seizures) but did not note an impact of KD on cerebral hemodynamics.	Plasma KB levels of βHB and acetoacetate were measured. Study noted a significant increase of plasma KBs during the study period, with βHB levels rising from 0.24 ± 0.31 mmol/l to 0.61 ± 0.53 mmol/l (p = 0.0005) and acetoacetate levels increasing from 0.19 ± 0.16 mmol/l to 0.52 ± 0.40 mmol/l (p < 0.0001) over the study period. Sustained reduction of plasma glucose from day one until day six.	GI side effects, noted an increase in several hepatocellular and cholestatic enzymes