Hello! This is Brandon from the Clinical Team, and I apologize for the delayed reply as I gathered my thoughts on this. Thank you for this thought-provoking question! I think it could probably be succinctly responded to with a “no, a permanent ketogenic diet (KD) is not required,” but certainly some further context is warranted 
.
A permanent KD might be necessary in a few instances. The first is Type 2 diabetes (T2D). Suppose a person is diagnosed with T2D and would like to put it into remission permanently. This would coincide with the need (or desire) to eliminate T2D-specific medications and ameliorate all signs and symptoms of the disease. In that case, a low-carb/KD diet is the nutrition you should follow. Strong evidence shows carbohydrate restriction can induce diabetes remission, often superior to standard care. This is achieved through reduced glycemic load, improved insulin sensitivity, weight loss, hepatic fat reduction, and beta-cell rest. One study showed 60% of patients achieved diabetes reversal (HbA1c < 6.5%, off diabetes meds except metformin) at 1 year on a well-formulated ketogenic diet (WFKD) [1]. The same cohort showed that 53.5% maintained diabetes reversal at year 2 [2], and again at year 5 [3].
Proof of concept: What happens if you reintroduce carbohydrates to this type of population? This is less frequently studied directly (because ethically, why would we want to make people get sick all over again?) but is known clinically and supported mechanistically and in trials with structured refeeding or diet discontinuation. When carbohydrate load increases, there are more glycemic excursions, leading to hepatic insulin resistance, beta-cell stress, and glycemic relapse. One study showed that 40% of initial remitters returned to diabetic status over 2 years with the reintroduction of carbohydrate intake [4].
The second instance where a KD permanency might be warranted is that there is convincing evidence that a KD can help with neurological diseases. The most extensive data from the early 1900s includes epilepsy, for which the KD is a great alternative or adjuvant to aggressive pharmacological options, which often bear significant side effects and marginal benefits for the disease.
If you seek neuroprotection, which we all should, then upregulating an alternative fuel source, like ketones, will help you bridge the energy gap. The said energy gap does appear to be age-related, but it is more pronounced in association with cognitive decline and Alzheimer’s Disease. Figure 1 below demonstrates how ketones (red) make up this energy gap [5].
Fig. 1 Causes and consequences of the brain energy gap in neurodegenerative disorders [5]
If not at risk, if “young”, and if metabolically flexible, indefinite KD may not be necessary. My colleague Dr. Rich LaFountain and I recently took a deeper dive into the concept of metabolic flexibility on a podcast, which will soon be posted here (still working on the post-editing!). However, metabolic flexibility is the ability of an organism (at the cellular and systemic level) to adapt fuel oxidation to fuel availability, efficiently switching between glucose and fatty acid metabolism depending on nutritional state, energy demand, and hormonal cues. In the fed state, insulin promotes glucose uptake and oxidation, suppressing lipolysis. In the fasted state, insulin levels drop, lipolysis and fatty acid oxidation increase, and gluconeogenesis is activated. Metabolic flexibility thus allows the dynamic switching between these pathways. To bring this to the table (pun intended), the general gist of metabolic flexibility is that there is a benefit to allowing some carbs into your diet to ensure you are not over-indexing in any one direction, as maintaining metabolic flexibility is important across the lifespan.
Many find long-term adherence to a KD challenging, and (if you ask me) I would more so assign blame to the availability of food and the social gathering component than any type of “will power”, hormonal, or physiological barrier. With that said, there are ways to increase ketone availability sans strict KD adherence, which include MCT and exogenous ketone supplements that can be helpful and do not require a habitual keto diet. This would ultimately be a personal decision, but the research is pretty clear that when an energy gap or deficit contributes to the etiology of neurodegenerative disease or declines, allowing for seamless fluctuation between alternative fuel sources, such as ketones, will provide some protection.
In the context of aging or individual shifts (reduction) in insulin sensitivity over time, it is more important to consider a foundational nutrition approach that emphasizes lower carbohydrate intake and flirts with the KD. Still, it is unlikely to be broadly applied across the adult lifespan. We don’t have direct studies here, but there is an apparent reduction in energy production capacity with age, so ketones increase in importance to aid in that, perhaps into the sixth decade of life and beyond. A good call to action here would be the importance of your own lab data, which would be a good way to determine when and where that is most appropriate. (Looking at markers such as fasting glucose, HbA1c, insulin, HOMA-IR, Triglycerides, HDL, LDL, ApoB, hsCRP, and even waist circumference and blood pressure).
One final point that is often overlooked is a deeper understanding of carbohydrates/glucose’s purpose beyond just being a 4kcal/g fuel source, especially in the context of a low-carb/KD diet. Even under low-carb conditions, glucose is still essential for specific cellular functions. The body shifts to conserve, recycle, and prioritize glucose for non-substitutable roles. Even in deep ketosis, glucose is indispensable for nucleotide synthesis (DNA/RNA), NADPH production (PPP: glutathione and lipid biosynthesis), ATP in glycolysis-dependent cells, glycosylation & protein modification, and gluconeogenesis cycling. Everything else shifts to fatty acids (β-oxidation), ketones (βHB, AcAc), and amino acids (in gluconeogenesis).
So, if you want an apple, eat an apple 
Sources:
[1] Effectiveness and Safety of a Novel Care Model for the Management of Type 2 Diabetes at 1 Year: An Open-Label, Non-Randomized, Controlled Study - PubMed
[2] Long-Term Effects of a Novel Continuous Remote Care Intervention Including Nutritional Ketosis for the Management of Type 2 Diabetes: A 2-Year Non-randomized Clinical Trial - PubMed
[3] 5-Year effects of a novel continuous remote care model with carbohydrate-restricted nutrition therapy including nutritional ketosis in type 2 diabetes: An extension study - PubMed
[4] https://www.thelancet.com/journals/landia/article/PIIS2213-8587(19)30068-3/abstract
[5] Brain energy rescue: an emerging therapeutic concept for neurodegenerative disorders of ageing - PMC
