May 5, 2021

Are Low-Carb Diets in Diabetics Worth the Effort?

Results from a randomized clinical trial
Type 2 diabetics following a low-carbohydrate diet showed improved glycemic control, but the long-term effectiveness requires further investigation.


Chen CY, Huang WS, Chen HC, et al. Effect of a 90 g/day low-carbohydrate diet on glycaemic control, small, dense low-density lipoprotein and carotid intima-media thickness in type 2 diabetic patients: an 18-month randomised controlled trial. PLoS One. 2020;15(10):e0240158.

Study Objective

To compare the effects of a 90-g/day low-carbohydrate diet (LCD) to those of a traditional diabetic diet (TDD) on glycemic control, lipids, weight, medication use, and atherosclerosis in type 2 diabetic patients.


A randomized, controlled study at the Department of Family Medicine, National Taiwan University Hospital, in Taiwan.


The trial recruited 92 participants with poorly managed type 2 diabetes, aged 20 to 80 years, with a normal or significantly elevated body mass index (BMI); however, only 85 patients completed the study.

Inclusion factors included having a diagnosis of diabetes for more than 1 year and poorly controlled hemoglobin A1c (HbA1c) levels greater than 7.5% 3 months prior to being randomly assigned to 1 of 2 groups: 90 g/day LCD (n=43) or TDD (n=42), regardless of medication use.

Participant demographics included information regarding highest level of education, sex, history of smoking, alcohol usage, family history of diabetes, diabetic treatment, and past medical history of hypertension.

Exclusion criteria from the trial included pregnancy, lactating women, impaired liver function, abnormal kidney function with serum creatinine greater than 1.5 mg/dL, liver cirrhosis, cardiovascular disease, frequent attacks of gout greater than 3 times per year, involvement in other weight-loss programs or the use of drugs for weight loss, eating disorders, and inability to successfully complete the questionnaire.

Study Parameters Assessed

For both groups, researchers implemented dietary intervention and surveillance. For the LCD, carbohydrate intake was 90 grams per day, divided into 6 servings of carbohydrates, and for the TDD group, researchers calculated total calorie intake per day by multiplying ideal weight by 25 kcal/kg for participants with a BMI between 18.5 and 24; ideal weight by 20 kcal/kg for overweight subjects with a BMI >24; and ideal weight by 30 kcal/kg for underweight subjects with a BMI of <18.5. Every 6 months, a blind evaluator took a 3-day weighted food record, using the E-Kitchen nutritional analysis software, to calculate the caloric and nutritional intake for enrollment numbers only.

Researchers assessed the use of medication and changes acquired during the trial using the medication effect score (MES), which evaluated the overall utilization of antidiabetic agents calculated by potency and dosage of the medication. The percentage of max dose for the medications were multiplied by an adjustment factor, and the total was summed up to produce the final MES value. Patients confirmed MES values at every visit to confirm their actual use. Researchers characterized the antidiabetic agents based on their mechanism and reported them within different categories along with the number of diabetic medications at the visits (number of tablets and number of insulin shots per day).

With the array of dietary choices, the best diet is one that allows the patient to have the most adherence based on the patient’s meal preferences, health status, and energy needs.

Researchers assessed physical activity using the International Physical Activity Questionnaire, Taiwan (IPAQ-Taiwan) every 3 months, and they adjusted medications every 6 months if HbA1c values were found to be higher than 8% or lower than 6.5%. A research assistant measured weight, blood pressure, BMI, body composition, and waist, hip, and thigh girths every 3 months. The assistant measured body composition using a Tanita Body Composition Analyzer BC-418.

Researchers attained blood samples after fasting to assess fasting glucose levels, serum lipids (total cholesterol, low-density lipoprotein [LDL], high-density lipoprotein [HDL], triglyceride, small dense LDL [sdLDL]), HbA1c, 2-hour blood samples for 2-hour glucose levels, and serum creatinine at every visit. Researchers assayed measurements of sdLDL with the sdLDL-Seiken kit at baseline and again at 6, 12, and 18 months.

They analyzed the microalbumin/creatinine ratio using random urine samples collected from patients at baseline and at 18 months; additionally, they also evaluated complete blood cell count, alanine aminotransferase (ALT), and uric acid during this time. Researchers measured carotid intima-media thickness (IMT) at baseline and at 18 months with 3 measurements taken from each side to calculate the average value of carotid IMT. They did not perform the measurement if there was a focal elevation of more than 5mm next to the point or intimal thickness of more than 15mm.

Primary Outcome Measures

Primary outcome measurements included glycemic control (HbA1c, fasting glucose, and 2-hour glucose) and change in MES. Additionally, secondary outcomes included lipid profiles, sdLDL, serum creatinine, microalbuminuria, and carotid IMT.

There were no significant changes from baseline after the completion of the trial for lipid profiles, sdLDL, serum creatinine, microalbuminuria, ALT, and carotid IMT between the groups. A mean difference at the end of the study occurred with a reduction of systolic blood pressure (SBP) and diastolic blood pressure (DBP) when compared to baseline for the LCD group (P<0.05) than for the TDD group (P<0.05), for which higher values were noted.

Daily total energy had no significant differences between the LCD and TDD groups except at time points of 6, 12, and 18 months (P<0.05). At the end of the 18-month trial, the groups consumed the following values of carbohydrates respectively: LCD=88.0±29.9 g and TDD=151.1±29.8 g (P<0.05). Significant changes were indicted for HbA1c (LCD=–1.6±0.3%, TDD=–1.0±0.3%), 2-hour glucose (LCD=–94.4±20.8 mg/dL, TDD=–18.7±25.7 mg/dL), MES (LCD=–0.42±0.32, TDD=–0.05±0.24), weight (LCD=–2.8±1.8 kg, TDD=–0.7±0.7 kg), waist circumference (LCD=–5.7±2.7 cm, TDD=–1.9± 1.4 cm), hip circumference (LCD=–6.1±1.8 cm, TDD=–2.9±1.7 cm), and blood pressure (LCD=–8.3±4.6/-5.0±3 mmHg, TDD=1.6±0.5/2.5±1.6 mmHg).

Daily protein and fat intake was significantly higher in the LCD group compared to the TDD group at 12 and 18 months (P<0.05). The mean difference in protein intake was significantly higher for the LCD group at 12 and 18 months compared to baseline, but not for the TDD group (P<0.05). The LCD group consumed more saturated and monosaturated fat (P<0.05) than the TDD group (P<0.05). Specifically, the LCD group consumed the highest amount of monosaturated fat at 18 months, while the TDD group consumed significantly lowered monosaturated fat at 6 months.

Key Findings

The clinical trial for this study had a 90% completion rate, due to these patients being treated routinely by their family physicians for chronic diseases before enrolling in the study. It was found that the duration of diabetes mellitus (DM) for the TDD group was 9.7±8.0 years and 10.1±7.8 years for the LCD group, respectively.

Those in the LCD group showed improved glycemic control, reducing MES, lowering blood pressure, and decreasing weight, hip, and waist circumference without causing adverse effects on lipid profiles, sdLDL, ALT, serum creatinine, microalbuminuria, and carotid IMT compared to the TDD group. HbA1c was reduced by twice the amount at 6 months, and the lowered value was maintained over the course of the 18-month period. Not only was the dietary effect of LCD beneficial in reducing HbA1c, but it also decreased the number of medications required, with a significant reduction from baseline MES within the group (P<0.05) and between the groups (P<0.05).

The increased intake of protein and saturated fat, in the LCD group had no adverse effects on LDL and sdLDL. sdLDL was found to not have any reduction in this study with LCD, as previously reported in nondiabetic and overweight patients with hyperlipidemia. Although the LCD group had vastly lowered triglycerides and increased HDL, the effects were not significantly different between the TDD and LCD groups.

There was a net weight reduction of 3 kg in the LCD group; however, the 18-month mean difference from baseline showed a substantial reduction in weight and waist and hip circumference for the LCD group when compared to the TDD group (P<0.05). Additionally there were lowered blood pressure values between the groups (P<0.05) and regulated ALT in the LCD group. This allowed the LCD group to attain further weight reduction and modified body image beyond the dietary theory of reduced caloric intake. Although SBP and DBP were significantly reduced from baseline for the LCD, the TDD group had increased levels, indicating a statistically significant (P<0.05) difference not observed in prior studies.

This study conducted the first examination of the modification in carotid IMT after moderate LCD intake and the effect LCD has on atherosclerosis versus a TDD. The end results were not very different; however, the LCD group had no change in their carotid IMT, and the TDD group had slightly increased values over the 18-month period.

Practice Implications

Diabetes has become a major health concern worldwide, with increased prevalence impacting morbidity and mortality globally.1 According to the Centers for Disease Control (CDC) National Diabetes Statistics Report of 2020, 34.2 million Americans (just over 1 in 10) have diabetes and 1 in 3 are prediabetic.2 The total estimated costs associated with the condition in the US in 2012 were $245 billion due to associated medical expenses and lower worker efficacy.1 However, the American Diabetes Association (ADA) estimates that the total costs of diagnosed diabetes have risen to $327 billion in 2017, according to new research released in 2018.3 The most common form of diabetes mellitus is type 2 (DM2) accounting for over 90% of incident diabetes. DM2 is unique from type 1 in that lifestyle choices are a large contributor to manifestation of the disease.1 Obesity and being overweight increase the risk for developing DM2. There are also several other factors that increase the risk of the development of DM2, such as environmental factors, genetic predisposition, social factors, and lifestyle factors.

Poorly controlled DM2 can affect the function of several organs in the body, increasing the risks for the progression of complications, including neuropathy, retinopathy, and kidney dysfunction.1 Inflammation is associated with the pathologies induced by DM2 due to insulin resistance.4 Activated immune cells enter into the tissues of the body, triggering inflammation. High glucose and increased fatty acids promote the release of a proinflammatory cytokine called innate immune-cell activator toll-like receptor 4 (TLR4), which is triggered by elevations of surface protein expression.4 Therefore, targeting inflammation may be a worthwhile strategy while pursuing better glucose control.5

Consuming foods with a higher glycemic index increases the risk of developing DM2 compared to foods with a lower glycemic index.1 The Food and Agriculture Organization, a United Nations agency, has recommended the use of the glycemic index as a guide for those with DM2.1 The glycemic index is a rough gauge of the relative effect of a given foodstuff on blood glucose, with higher values reflecting higher glucose-raising effects. Besides the effect on glucose, some carbohydrates can be particularly harmful. For example, fructose exerts harmful effects on the liver, causing increased fat accumulation and hastening steatohepatitis.6

Patients with a long history of poorly controlled diabetes are difficult to treat, as shown by previous studies.7 With the increased prevalence of obesity and DM2, effective treatment options are needed to help with managing health issues of those affected. Lifestyle therapy is an ideal treatment suggestion for improving glucose control in people with DM2.4 In treating patients with DM2, nutritional management is essential, as dietary changes can help control glucose, manage weight, and reduce other associated health risks. The World Health Organization (WHO) has recommended limiting daily intake of carbohydrates to reduce the risks of obesity, diabetes, and cardiovascular diseases.8

Most current dietary guidelines advocate for eating a clean and healthy diet low in saturated fat, with an emphasis on low-glycemic index foods.4 However, evidence increasingly shows that a low-carbohydrate, high-fat diet is more effective for lowering glucose than simply limiting carbohydrate intake.4 Most diabetic organizations recommend a traditional diabetic diet (TDD) with protein intake of 1.0–1.2 g/kg, carbohydrate intake of 50% to 60%, and a total fat of ≤30% of total caloric intake.7 However, there is compelling evidence indicating the effectiveness of a low-carbohydrate diet (LCD) for weight loss and glycemic control of DM2.7

This current study under review shows that a 90-g/day LCD, high in fat and with moderate protein intake, resulted in better glycemic control without opposing effects on cardiovascular health, making it a practical dietary suggestion for type 2 diabetes.7 A diet that causes reduced energy intake is more favorable for weight loss and related improved metabolic and functional changes.8

LCD can provide short-term improvements in type 2 diabetic patients by lowering cardiovascular risk, allowing glycemic control, and promoting weight loss.7 A very low-carbohydrate intake (< 50 g/d) will result in lowered glucose supply to areas such as the muscles, brain, and liver, resulting in a decline of the glucose stored as glycogen.8 In patients with DM2, consuming an LCD helped them to have a feeling of satiety by having less desire to eat.9 A prior LCD study of a moderate 130-g/d diet showed effectiveness at 6 months, but 2 earlier studies showed no further effects of the LCD at a 1- and 2-year follow-up after completion of the study, indicating reduced effectiveness long-term.7 More research is needed to determine the effectiveness of implementing an LCD for a longer period of time since there are theoretical concerns for other potential health risks.

Another potential reason why the LCD loses efficacy long-term is due to the reduction in glycogen stores associated with low-carbohydrate diets, which can result in reduced physical activity.10 Long-term LCD can have potential adverse effects on lipids in the body due to variable responses on total blood cholesterol and LDL concentrations to LCD.7 According to the findings of Halton et al, the LCD, despite having adverse effects on total cholesterol and LDL cholesterol levels, had a beneficial effect on HDL cholesterol and triglyceride levels.11

According to a previous study, between the LCD and TDD groups, there was an estimated absolute HbA1c reduction of 0.5%, with a standard deviation of 0.408%.7 Similarly, a paleolithic diet typically resembles a low-carbohydrate diet, with the main focus on eating fresh, healthy foods and removing processed foods.12 A paleo diet is lower in carbohydrates, higher in protein, and moderate or higher in fat.12 The paleolithic diet resulted in lower HbA1c, triglycerides, DBP, and weight and waist circumference, and higher HDL, when compared to a standard diabetes diet.13 A study using the paleolithic diet in diabetics found the diet to improve glycemic control in the participants, with reduction of HbA1c levels by –0.4 percentage points lower than the diabetes diet.13

Improved insulin functions allow for less medication needed to support insulin metabolism, and this decreases the need for diabetes medication. Based on a randomized trial of a LCD conducted by Saslow et al, an improvement in glycemic control occurred despite the significant decreases in diabetes medications, particularly sulfonylureas, in the low-carbohydrate group. The LCD was also more effective than a moderate carbohydrate diet at reducing HbA1c.14 The study under review found similar effects, with a significant reduction in diabetic medications in the LCD vs TDD group. According to one study, the estimated absolute MES reduction between the LCD and TDD groups is 0.4, with a standard deviation of 0.5.7

Cardiovascular disease is a major concern for individuals with DM2. The leading cause of morbidity and mortality in diabetic patients is atherosclerosis.7 Modified sdLDL can induce inflammatory processes associated with cardiovascular disease.15 Diabetes mellitus is associated with sdLDL particles, which increase risks for heart disease.7 A smaller prospective study conducted on type 2 diabetic and prediabetic patients demonstrated that sdLDL was predictive of insulin resistance.15 Carotid IMT is a marker for atherosclerosis used worldwide, and it can be used as a predictor for future cardiovascular events.16 Carotid IMT increases in patients with adult DM2 due to impaired glucose metabolism as a result of atherosclerotic processes occurring in the arteries.17 Evidence suggests that a LCD may reduce the number of sdLDL particles in obese and nondiabetic hyperlipidemia patients; however, the sdLDL in diabetic patients needs to be investigated.7

Dietary treatment is necessary for the management of diabetic patients. More research is also needed to determine the best course of dietary treatment for these patients, as well as patients suffering from obesity, for whom there is no optimal dietary plan to facilitate long-term health management.18 With the array of dietary choices, the best diet is one that allows the patient to have the most adherence based on the patient’s meal preferences, health status, and energy needs.18 Patients who are taking multiple medications, have dietary restrictions, or have other existing comorbidities should consult a licensed healthcare professional or dietitian before beginning a low-carbohydrate diet to facilitate optimal nutritional intake.6

Patient compliance is highly important as patients can revert to their previous diet over time, so it’s important to teach patients skills for maintaining healthy changes. However, the best nutritional treatment strategy for prevention and treatment of DM2 is one that is ultimately individualized, taking into account factors such as culture, food availability, and patient preference, but encourages consuming a variety of the foods from the main food groups each day.19

Categorized Under


  1. Ojo O, Ojo OO, Adebowale F, Wang XH. The effect of dietary glycaemic index on glycaemia in patients with type 2 diabetes: a systematic review and meta-analysis of randomized controlled trials. Nutrients. 2018;10(3):373.
  2. National Diabetes Statistics Report, 2020. Centers for Disease Control Prevention Web site.,1%20in%203%E2%80%94have%20prediabetes. Accessed March 5, 2021.
  3. The cost of diabetes. American Diabetes Association Web site.,in%20the%20absence%20of%20diabetes. Accessed March 5, 2021.
  4. Myette-Côté É, Durrer C, Neudorf H, et al. The effect of a short-term low-carbohydrate, high-fat diet with or without postmeal walks on glycemic control and inflammation in type 2 diabetes: a randomized trial. Am J Physiol Regul Integr Comp Physiol. 2018;315(6):R1210-R1219.
  5. Dandona P, Aljada A, Bandyopadhyay A. Inflammation: the link between insulin resistance, obesity and diabetes. Trends Immunol. 2004;25(1):4-7.
  6. Kelly T, Unwin D, Finucane F. Low-carbohydrate diets in the management of obesity and type 2 diabetes: a review from clinicians using the approach in practice. Int J Environ Res Public Health. 2020;17(7):2557.
  7. Chen CY, Huang WS, Chen HC, et al. Effect of a 90 g/day low-carbohydrate diet on glycaemic control, small, dense low-density lipoprotein and carotid intima-media thickness in type 2 diabetic patients: an 18-month randomised controlled trial. PLoS One. 2020;15(10):e0240158..
  8. Brouns F. Overweight and diabetes prevention: is a low-carbohydrate-high-fat diet recommendable? [published correction appears in Eur J Nutr. 2019 Apr 16]. Eur J Nutr. 2018;57(4):1301-1312.
  9. Wang LL, Wang Q, Hong Y, et al. The effect of low-carbohydrate diet on glycemic control in patients with type 2 diabetes mellitus. Nutrients. 2018;10(6):661.
  10. Winwood-Smith HS, Franklin CE, White CR. Low-carbohydrate diet induces metabolic depression: a possible mechanism to conserve glycogen. Am J Physiol Regul Integr Comp Physiol. 2017;313(4):R347-R356.
  11. Halton TL, Willett WC, Liu S, et al. Low-carbohydrate-diet score and the risk of coronary heart disease in women. N Engl J Med. 2006;355(19):1991-2002.
  12. Editor. The paleolithic (or paleo) diet is based on the food that is believed to be similar to the daily diet of cave people. Web site. Accessed March 5, 2021.
  13. Jönsson T, Granfeldt Y, Ahrén B, et al. Beneficial effects of a paleolithic diet on cardiovascular risk factors in type 2 diabetes: a randomized cross-over pilot study. Cardiovasc Diabetol. 2009;8:35.
  14. Saslow LR, Kim S, Daubenmier JJ, et al. A randomized pilot trial of a moderate carbohydrate diet compared to a very low carbohydrate diet in overweight or obese individuals with type 2 diabetes mellitus or prediabetes. PLoS One. 2014;9(4):e91027.
  15. Ivanova EA, Myasoedova VA, Melnichenko AA, Grechko AV, Orekhov AN. Small dense low-density lipoprotein as biomarker for atherosclerotic diseases. Oxid Med Cell Longev. 2017;2017:1273042.
  16. Nezu T, Hosomi N, Aoki S, Matsumoto M. Carotid intima-media thickness for atherosclerosis. J Atheroscler Thromb. 2016;23(1):18-31.
  17. Bulut A, Avci B. Carotid intima-media thickness values are significantly higher in patients with prediabetes compared to normal glucose metabolism. Medicine (Baltimore). 2019;98(44):e17805.
  18. Sandouk Z, Lansang MC. Diabetes with obesity—Is there an ideal diet? Cleve Clin J Med. 2017;84(7 Suppl 1):S4-S14.
  19. Deed G, Barlow J, Kawol D, Kilov G, Sharma A, Hwa LY. Diet and diabetes. Aust Fam Physician. 2015;44(5):192-196.