Currently, treatment of type 1 diabetes mellitus (T1DM) relies on insulin replacement together with a high-carbohydrate diet [1]. Two group studies, however, showed that putting patients on a low-carbohydrate diet results in decreased need for insulin, lowers the number of hypoglycemic episodes and can be maintained on the long-term without side-effects [2][3]. The classical ketogenic diet is an established and effective treatment in epilepsy [4]. We are aware of four case studies in literature where epilepsy co-occurring with T1DM was treated with the classical ketogenic diet [5][6][7][8]. Recently, we have published a case of a 19-year-old patient with T1DM who was successfully managed with a modified version of the ketogenic diet we refer to as the Paleolithic ketogenic diet [9]. In that case insulin could be discontinued, glucose levels normalized and the level of C-peptide increased more than three-fold within two months. Herein, we report a case of a child with T1DM who was successfully treated with the Paleolithic ketogenic diet. The child has a follow-up of 19 months.

Medical history
The nine-year-old child had a three-month history of polyuria, polydipsia, polyphagia, fatigue and weight loss. His abdomen was bloated. He had frequent upper respiratory tract infections, epistaxis and eczema on his hands as well as asthma. A laboratory workup on 01 January 2014 showed glucose level of 692 mg/dl and glucose in the urine. A laboratory test on 07 January 2014 showed mild positivity of anti GAD (glutamic acid decarboxylase) antibodies (1280 IU/ml). He was placed on an insulin regime (21 IU daily) with six meals containing 210 g carbohydrate. He was on the conventional insulin therapy for six weeks. Although he strongly adhered to the prescribed diabetes diet his glucose levels fluctuated largely and his complaints (fatigue, eczema, bloated abdomen) persisted too. Therefore, parents of the child sought for alternative treatment options.

Paleolithic ketogenic diet
We first met the child and his parents on 13 February 2014. We advised the Paleolithic ketogenic diet consisting of only animal meat, fat, offal and eggs with a fat:protein ratio of about 2:1. Following our advice the child consumed meat, fat and offal only from cattle and pork. He did not consume vegetables and fruits at all. Artificial sweeteners and vegetable oils were excluded. He was allowed to consume small amounts of honey. The diet was introduced gradually within five days. During this time insulin was adjusted according to fingerstick glucose measurements. From the sixth day (from 18 February 2014), the child was on a full Paleolithic ketogenic diet. C-peptide measurement was performed on 13 February 2014 which showed 1.1 ng/ml (normal range: 0.9–7.1 ng/ml). Given that on the Paleolithic ketogenic diet glucose levels were normal both before and after meals insulin was discontinued. The child was advised to eat when hungry. Typically, he had three meals a day. Ketosis was regularly checked by urinary ketone strips which showed sustained ketosis. Food records together with fingerstick glucose were tracked daily. This latter showed normal levels both preprandially and postprandially. For statistical calculations and visualization purposes we selected fingerstick glucose measurements for the meal closest the noon (Figure 1). Glucose levels were significantly lower during the Paleolithic ketogenic diet as compared to the six weeks of the insulin regime both preprandially (Paleolithic ketogenic diet mean: 75.7 mg/dl; insulin regime mean: 146 mg/dl) and postprandially (Paleolithic ketogenic diet mean: 93.7 mg/dl; insulin regime mean: 198.2 mg/dl). Glucose levels showed higher fluctuations during the insulin regime as also shown by larger standard deviation of glucose values on the insulin regime (SD=72 mg/dl) as compared to the Paleolithic ketogenic diet (SD=15.5 mg/dl). Hypoglycemic episodes which were frequent on the standard insulin regime completely disappeared while on the Paleolithic ketogenic diet. Currently, the child is on the Paleolithic ketogenic diet for 19 months without interruption (Figure 1).

Laboratory workup
Laboratory workups were carried out six times (Table 1). While on the Paleolithic ketogenic diet glucose levels on laboratory tests were between 86 and 130 mg/dl whereas his HgA1c levels were between 5.2 and 5.6%. Cholesterol and LDL cholesterol were slightly elevated. Uric acid was normal. Triglyceride measurements were normal except for elevations on 19 Junuary 2014 and 03 February 2015 which were associated with parent reports of increased food intake. Other laboratory blood parameters were normal. Urinary ketones were positive on each laboratory test. C-peptide measurements indicated 0.7 ng/ml on 15 September 2014 and 0.5 ng/ml on 09 February 2015. Thus measured C-peptide levels seemed to decrease over time. The decrease of C-peptide levels following diagnosis onset is a well-known phenomenon in T1DM and is attributed to the progression of the autoimmune destruction of insulin secreting beta cells [10]. In our case, however, the very low carbohydrate intake might also have resulted in decreased levels of C-peptide. In order to get clues for distinguishing between the two possibilities we measured C-peptide twice two days apart at the same time of the day (in the morning). The first measurement was carried out in fasting (thus after a skipped breakfast) while the second one was carried out following having a usual breakfast. In other words, in the second case we measured a stimulated C-peptide, however, stimulation was done without carbohydrates only using his regular Paleolithic ketogenic diet. Fasting C-peptide on 27 April 2015 was 0.5 ng/ml while 1.2 ng/ml on 29 April 2015 following stimulation (Figure 2). Given that no data are available on C-peptide levels on a low carbohydrate diet in healthy subjects we additionally had two healthy normal controls (females, aged 39 and 42 years, respectively) who had been following the Paleolithic ketogenic diet to measure their fasting C-peptide levels. The two subjects were instructed to consume no carbohydrate for three days prior to blood test. These subjects had a C-peptide level of 1 and 1.3 ng/ml thus levels which also fall around the lower limit of the normal range.

Continuous subcutaneous glucose monitoring
We additionally performed continuous subcutaneous glucose monitoring (Medtronic, Guardian Real-Time Continuous Glucose Monitoring System) for six days between 22 and 28 April 2015. The system provides an average tissue glucose measurement every five minutes. Sensor daily overlay is shown in Figure 2. On average glucose level was of 101 mg/dl. Glucose levels did not substantially change following having meals: during the 60 minutes before meals glucose level was 97 mg/dl on average while during the 60 minutes following meals it was of 94 mg/dl on average (Figure 3).

Changes in clinical and physical features
Shortly, after diet onset the child reported increased physical and mental fitness. His eczema disappeared within a few weeks. While on the paleolithic ketogenic diet he reported no epistaxis and a great reduction in his upper respiratory infections and associated symptoms. For the change in his physical appearance, including the disappearance of his bloated abdomen and developmental achievements (Figure 4). At the time of diagnosis he weighted 24 kg and his height was 134 cm (BMI=13.37). Currently, (19 months after diet onset) he weights 28.4 kg and his height is 140 cm (BMI=14.49).

Informed consent
Parents of the child gave written informed consent for the publication of this case.

The child strictly adhered to the Paleolithic ketogenic diet as assessed by home monitoring of glucose, urinary ketones, laboratory workups and frequent feedback from the patient and his parents. They reported that maintaining the diet was relatively easy for him. Dietary adherence resulted in normalized glucose levels which allowed insulin freedom. Measurement of the stimulated C-peptide indicated normal insulin secretion.

Our experience indicate that in new-onset T1DM the Paleolithic ketogenic diet results in lowered insulin need which may be covered by the residual insulin secretion in these patients [9]. Honeymoon effect shortly after diagnosis onset is frequently described in T1DM [11]. However, normoglycemia and insulin-freedom maintained across 19 months in our patient may indicate that the patient is well beyond the honeymoon period and that the autoimmune process destructing the insulin secreting beta cells of the pancreas was halted or reversed. This may have been due to the restriction of non-Paleolithic substances, including milk and dairy, which may promote inflammation and autoimmune processes [12].

Neither side effects nor complications emerged while on the diet. Rather the Paleolithic ketogenic diet was beneficial in several respects. The child's asthma, eczema, epistaxis, bloated abdomen, frequent upper respiratory infections and fatigue disappeared. Positive effects are in line with those presented in our other patients put on the Paleolithic ketogenic diet [13] [14][15][16]. Although we did not use vitamin or other supplements no deficiencies emerged.

In literature, this is the second case report of T1DM treated with the Paleolithic ketogenic diet without insulin. In both cases the Paleolithic ketogenic diet was initiated in an early period of the clinical manifestation of the disease. Although there are indications that insulin requirement may be lowered on diets low in carbohydrate [2][3] we are not aware of any dietary intervention or other treatment modalities where insulin replacement could be discontinued in T1DM.

We opine that the Paleolithic ketogenic diet ensure normal glucose levels and can be maintained on the long-term in those patients with newly diagnosed T1DM with residual insulin secretion. It is important to emphasize, however, that in those patients with long-standing T1DM beta cells might have exhausted and therefore there may be a need for insulin replacement. In these cases, however, the Paleolithic ketogenic diet may be used as an adjunct in an attempt to likely prevent diabetic complications.

1. Chiang JL, Kirkman MS, Laffel LM, Peters AL; Type 1 Diabetes Sourcebook Authors. Type 1 diabetes through the life span: a position statement of the American Diabetes Association. Diabetes Care 2014 Jul;37(7):2034–54.   [CrossRef]   [Pubmed]    
2. Nielsen JV, Jönsson E, Ivarsson A. A low carbohydrate diet in type 1 diabetes: clinical experience--a brief report. Ups J Med Sci 2005;110(3):267–73.   [CrossRef]   [Pubmed]    
3. Nielsen JV, Gando C, Joensson E, Paulsson C. Low carbohydrate diet in type 1 diabetes, long-term improvement and adherence: A clinical audit. Diabetol Metab Syndr 2012 May 31;4(1):23.   [CrossRef]   [Pubmed]    
4. Kossoff EH, Rho JM. Ketogenic diets: evidence for short- and long-term efficacy. Neurotherapeutics 2009 Apr;6(2):406–14.   [CrossRef]   [Pubmed]    
5. Henwood MJ, Thornton PS, Preis CM, Chee C, Grimberg A. Reconciling diabetes management and the ketogenic diet in a child with pyruvate dehydrogenase deficiency. J Child Neurol 2006 May;21(5):436–9.   [Pubmed]    
6. Dressler A, Reithofer E, Trimmel-Schwahofer P, et al. Type 1 diabetes and epilepsy: efficacy and safety of the ketogenic diet. Epilepsia 2010 Jun;51(6):1086–9.   [CrossRef]   [Pubmed]    
7. Aguirre Castaneda RL, Mack KJ, Lteif A. Successful treatment of type 1 diabetes and seizures with combined ketogenic diet and insulin. Pediatrics 2012 Feb;129(2):e511–4.   [CrossRef]   [Pubmed]    
8. Aylward NM, Shah N, Sellers EA. The ketogenic diet for the treatment of myoclonic astatic epilepsy in a child with type 1 diabetes mellitus. Can J Diabetes 2014 Aug;38(4):223–4.   [CrossRef]   [Pubmed]    
9. Tóth C, Clemens Z. Type 1 diabetes mellitus successfully managed with the paleolithic ketogenic diet. Int J Case Rep Images 2014;5(10):699–703.   [CrossRef]    
10. Wang L, Lovejoy NF, Faustman DL. Persistence of prolonged C-peptide production in type 1 diabetes as measured with an ultrasensitive C-peptide assay. Diabetes Care 2012 Mar;35(3):465–70.   [CrossRef]   [Pubmed]    
11. Abdul-Rasoul M, Habib H, Al-Khouly M. 'The honeymoon phase' in children with type 1 diabetes mellitus: frequency, duration, and influential factors. Pediatr Diabetes 2006 Apr;7(2):101–7.   [CrossRef]   [Pubmed]    
12. Cordain L, Eaton SB, Sebastian A, et al. Origins and evolution of the Western diet: health implications for the 21st century. Am J Clin Nutr 2005 Feb;81(2):341–54.   [Pubmed]    
13. Clemens Z, Kelemen A, Fogarasi A, Tóth C. Childhood absence epilepsy successfully treated with the paleolithic ketogenic diet. Neurol Ther 2013 Sep 21;2(1-2):71–6.   [CrossRef]   [Pubmed]    
14. Tóth C, Clemens Z. Successful treatment of a patient with obesity, type 2 diabetes and hypertension with the paleolithic ketogenic diet. Int J Case Rep Images 2015;6(3):161–7.   [CrossRef]    
15. Tóth C, Clemens Z. Gilbert's syndrome successfully treated with the paleolithic ketogenic diet. Am J Med Case Rep 2015;3(4):117–20.    
16. Clemens Z, Kelemen A, Tóth C. NREM-sleep associated epileptiform discharges disappeared following a shift toward the paleolithic ketogenic diet in a child with extensive cortical malformation. Am J Med Case Rep 2015;3(7):212–5.   [CrossRef]