A Preliminary Retrospective and Prospective Cohort Study on a Traditional Chinese Long-term Extreme Fasting

Long-term extreme fasting (Bigu regimen) Procedures

All subjects receiving Bigu regimen should stay together for at least the first three days with the supervision by an expert. Exercises and meditation should be performed twice a day (usually once in the morning and once in the afternoon). The whole process of Bigu regimen was divided into two sections. For the first phase, participants should be maintained in a water-only or extremely low-calorie intake diet for 21 days under the instruction of an experienced supervisor. The source of calorie was mainly from brown sugar water and honey water (∼141.8kJ), each of which should be taken less than 1000mL per day. Participants who suffered from severe starvation could have 4g of dates, 3g of peanuts and 2g of walnuts (∼138.1kJ). Comparison of macronutrient and micronutrient intake between this fasting procedure and normal diet was shown as Table 2. For the second phase, these participants resumed taking foods incrementally for the first seven days (from day 22 to day 28). A detailed recommended diet composition during the 7-day resuming period was listed as Table 3. Participants could not take normal diets until finishing the second phase in order to avoid accidental damage to the gastrointestinal tract.

Patient cohorts

This study (ClinicalTrials.gov; NCT03193177) contained three progressive sections (Table 1). We firstly took a basic survey on 124 people who voluntarily performed the Bigu regimen for 21 days. Of these 124 subjects, 53 participants were completely followed before, during and after Bigu regimen when their weight, blood pressure and blood biochemistry biomarkers were examined. According to these survey results, we prospectively followed 20 participants with aberrant biomarkers of cardiovascular diseases, and tracked their physical indications at pre-fasting, fasting and post-fasting. This study was conducted in Nanshan Branch of Qilu Hospital from June to September in the year 2017 and 20 subjects were voluntarily enrolled into this observational study. Of these 20 participants, 18 had succeeded in accomplishing the entire period of fasting, 1 withdrew on day 8 because of scheduling conflicts and 1 withdrew on day 14 due to the noncompliance of Bigu regimen. All participants have signed voluntary informed consent for collection and analysis of blood and fecal microbiome samples under Institutional Review Board (IRB)–approved protocols (THUMED-BG-170612). All clinical examination and evaluation was carried out independently and recorded instantaneously.

Statistical analysis

After verify normally distribution of the data, comparison between two time points (e.g. day 0 versus day 21) was performed using paired two-tailed Student’s t tests. When the data does not follow normal distribution, comparison was carried out using Wilcoxon signed-rank test. P values <0.05 were considered significant. *** means P<0.001, ** means 0.001<P<0.01 and * means 0.01<P<0.05.

Role of the funding source

The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

A basic survey of physical markers follows 124 bigu participants

From March 2016 to Oct 2017, we followed 124 people who carried out 21-day Bigu regimen and recorded their physical markers just pre-fasting (day 0), mid-fasting (day 10), end-of-fasting (day 21) and 10-day post fasting (day 31) (Figure 1A). Of these 124 participants, 48 (39%) were male and 76 (61%) were female. During fasting, the participants were asked to report any physical discomfort in time. The most common self-reported grade 1 symptom is fatigue, while dry mouth, tingling and nausea were also listed. Weakness, vomiting, asthma and fever has been reported occasionally as the grade 2 symptoms. No severe adverse event was reported as grade 3 and above (Figure 1B). We investigated the normal activities of each participant by using a detector MI Band worn on the wrist, to record walking distance, heart rates, sleeping time and consumed energy. The data presented no significant difference during fasting compared to those at day 0 (Figure 1C), indicating those participants kept daily activities during the procedure. These data suggested that the long-term extreme (21-day, <5% of normal diet) fasting procedure causes no adverse events beyond grade 2 and is likely be safe to most normal people.

• Download figure
• Open in new tab

Figure 1. A preliminary survey of physical biomarkers in 53 subjects during 21-day fasting procedure.

A. Schema of fasting procedure in this survey. B. Self-reported adverse effects were recorded according to CTCAE, where 0=No symptoms, 1=Mild, 2=Moderate, 3=Severe, 4=Life threatening, 5=Death. C. Daily activities were recorded using Mi Band. D. Basic physical biomarkers were examined during the entire fasting procedure.

A detailed analysis on physical biomarkers of 53 participants

We further reviewed the data of all 124 participants, 53 of which had serial physical indices. Among the 20 examined indices, the BMI, diastolic blood pressure and systolic blood pressure decreased significantly in participants during fasting, compared to day 0 (Figure 1D). Meanwhile, some metabolic biomarkers, including blood urea nitrogen (BUN), total cholesterol (TC), triglyceride (TG), low density lipoprotein (LDL), creatinine (CRE) and high density lipoprotein (HDL) also declined during this procedure (Figure 1D). No microelements fell below normal levels, though Fe, Ca, Cl and P showed slightly decreases (Figure 1D). Unexpectedly, the fasting glucose was not detected decrease on day 21 (end-of-fasting), ruling out hypoglycemia during this period (Figure 1D). These data indicated that the long-term extreme fasting did not cause severe deviations out of the normal ranges of those physical indices, though as expected, some biomarkers (e.g. uric acid, triglyceride) had a significant fluctuation. The results also presented that the fasting procedure could decrease the blood pressure and blood cholesterol associated biomarkers, while kept blood glucose balanced, implying its potential therapeutic effects on hypertension and hyperlipidemia.

A prospective study on 20 participants with hypertension or hyperlipidemia

In order to examine whether this long-term extreme fasting would benefit those people with hypertension and/or hyperlipidemia, we prospectively recruited 20 unhealthy people (in one Bigu camp) and detected their dynamic changes of metabolic physical biomarkers, from June 18^th to Sep 8^th in the year 2017. Among the 20 participants, 7 (35%) were male and 13 (65%) were female. Eighteen of 20 participants had succeeded in accomplishing the entire period of fasting, 1 withdrew on day 8 because of scheduling conflicts and 1 withdrew on day 14 due to the noncompliance of Bigu regimen (Figure 2A). Of note, we kept collecting data from those 18 participants until day 51 (30-day post resuming). Before fasting (examined on day 0), 16 participants were overweight, 11 suffered from hypertension, 9 were hyperlipidemia, 3 were hyperglycemia, and 9 had fatty liver (Table 4).

• Download figure
• Open in new tab

Figure 2. A deep investigation of physical biomarkers in 20 subjects with metabolic diseases.

A. Schema of fasting procedure in the clinical trial. B-E. Changes of non-invasive physical indexes (B), hypertension associated biomarkers (C), biochemical biomarkers (D) and hormone levels (E) were evaluated during the fasting procedure.

We following carried out physical examination and biomarker detection on day 0, day 4, day 7, day 14, day 21 and day 51 (Figure 2A). The data showed that the 21-day fasting resulted in a significant reduction by 9.2 ±3.2 kg (P < 0.001) of body weight, accordance with a reduction by 3.3 ±1.0 of BMI (Figure 2B). Systolic blood pressure was reduced by 28.7± 17.8 mmHg, whereas diastolic blood pressure did not reach a significant change. Levels of three markers associated with hypertension including rennin, angiotensin I and angiotensin II were detected significant increases (P < 0.001, Figure 2C), indicating a feedback to elevate blood pressure through the rennin-angiotensin system. We also found that both total body lean and total body fat were significantly decreased during fasting process (Figure 2B), indicating the consumption of both muscles and fats for energy production.

Since most of the participants recruited suffered from metabolic diseases, we followed their metabolism associated biomarkers during the fasting procedure (Figure 2D). In the whole period, the concentrations of Na, Cl, P, K and Ca were not found to deviate beyond their normal ranges, while they still experienced fluctuations. The data showed that the levels of alanine transaminase, aspartate transaminase, total bilirubin, indirect bilirubin, direct bilirubin, ferritin, lactate dehydrogenase and creatine kinase were increased significantly. As glucose metabolic pathway was inhibited due to the lack of carbohydrate intake, levels of total cholesterol and low density lipoprotein, as well as total protein were elevated transiently during this period, but the level of high density lipoprotein was found to be decreased. The level of uric acid was found to be significantly increased after starting fasting, which did not return to the normal level until resuming normal diet (Figure 2D). The glucose level was decreased during the first seven days of fasting, while it started to increase on day 14, which was beyond our expectations. However, the insulin level started to decrease just after starting fasting and remained the lower level during the whole procedure.

As hormone plays an important role in regulating physical activities, dynamics of the levels of thyroid hormones and sex hormones were recorded at indicated time points (Figure 2E). For thyroid hormones, both TSH and serum free T3 levels were decreased significantly during fasting, the level of T4 was increased transiently on day 14, but ATG level did not show an obvious change. But among sex hormones including luteinizing hormone, follicle-stimulating hormone, prolactin, estradiol, progesterone and testosterone, only the levels of follicle-stimulating hormone and luteinizing hormone in women significantly increased during fasting procedure and maintained at the higher levels even on day 51.

Then we evaluate the effects of the 21-day fasting on metabolic diseases by assessing the changes of biomarkers and imaging examinations (Table 5). We examined the indices of all participants on day 21. The data showed that among the 12 participants with higher BMI (>23.9), 6 were reduced to the normal range, and the other 6 had a significant decrease (P<0.001). Of 11 participants suffering from hypertension, 9 decreased to the normal range, and the remaining 2 also had a remarkable decrease. Besides, the long-time extreme fasting were proved to be effective on fatty liver. We found that in 9 participants with fatty liver, 7 were ameliorated, including 2 from severe to moderate or mild, 4 from moderate to mild, and 1 from mild to non-fatty liver (Figure S1A-I). Conclusively, these data suggested that bigu regimen could benefit the unhealthy people in the high BMI, hypertension and fatty liver.

Post hoc analysis of health-threatening factors

As abnormal levels of biomarkers leading to increased risk factors of metabolic diseases, we selected clinically relevant cutoffs and compared normal and at-risk subjects for each risk factor (Figure 3), where systolic blood pressure >140mmHg or diastolic blood pressure >90mmHg are associated with hypertension, total cholesterol >6mmol/L and LDL >3.3mmol/L are associated with hyperlipidemia and fasting glucose >6.1mmol/L is associated with hyperglycemia. The data showed that subjects with a systolic blood pressure >140mmHg experienced a greater reduction by 18.6±10.7mmHg in systolic blood pressure than those with normal systolic blood pressure by 3.4±12.4mmHg (P=0.039). However, comparison between participants with diastolic blood pressure >90mmHg and those with normal diastolic blood pressure did not show any difference. Total cholesterol was reduced significantly in participants with baseline levels above 6 mmol/L by 1.3±2.1 mmol/L, while it was slightly elevated in participants with normal baseline levels by 0.76±0.99 mmol/L (P=0.01 between groups). LDL was reduced by 0.34±0.68 mmol/L in those with LDL >3.3 mmol/L but was not reduced in normal-range participants (P=0.016 between groups). Besides, BMI changes between groups with normal range and overweight did not show significant difference (P=0.281). These data indicated that the 21-day fasting had more pronounced effects in subjects with aberrant levels of these health-threatening factors than in subjects with normal levels of these physical indices.

• Download figure
• Open in new tab

Figure 3. Post hoc analysis of physical biomarkers among subgroups classified by levels of aging/disease risk factors.

All the 20 participants were classified into two or three subgroups by the severity of aging/disease risk factors. The Δ value represents a comparison to the baseline. P values were indicated between different subgroups at the same time points.

Typical case reports

Case #17 was the oldest man among these participants (Table 5). He was a 79 years old and was diagnosed with severe fatty liver just before starting fasting. Then he performed the 21-day fasting with a complete caloric deprivation without any major inconvenience and lost 7.1 kilogram (from 80.6kg to 73.5kg). Levels of his blood pressure, total cholesterol and low-density lipoprotein did not change notably (Figure 4A), while the extent of his fatty liver had a great improvement from severe fatty liver to mild fatty liver on day 51 (Table 5, Figure S1F). Notably, he had not received any treatment for his fatty liver from day 0 to day 51 after starting fasting.

• Download figure
• Open in new tab

Figure 4. Dynamics of the levels of physical biomarkers for specific cases.

The data were examined at indicated time points. Dash areas show the normal ranges.

Case #4 was a 60-year-old Chinese women diagnosed with hypertension and mild-moderate fatty liver (Table 5, Figure S1A). Her BMI was 29.6, much higher than the upper limit value 23.9. Different from other participants, she suffered from varied adverse effects during fasting, including fatigue, weakness, muscle pain, vomiting, headache and sometimes sleep disorder. Her systolic blood pressure also transiently raised from 141mmHg (day 0) to 166mmHg on day 4, but then started to decline and reached to 116mmHg at the end of 21-day fasting (Figure 4B). Her hyperlipemia-associated biomarkers did not change significantly after fasting, though transient increases occurred on day 4, day 7 and day 14 (Figure 4B). After fasting, the extent of her fatty liver turned better from mild-moderate on day 0 to mild on day 51 (Table 5, Figure S1A) through B-sonic without any other treatment.

Case #10 was a 21-year-old Chinese man diagnosed with severe fatty liver, hypertension and overweight. His BMI was 44.3, the highest among all the participants, and his systolic blood pressure was 159 mmHg. He also experienced fatigue, weakness, nausea and vomiting occasionally during fasting. However, his BMI was decreased to 38.7 on day 21 and 39.3 on day 51; and his systolic blood pressure declined to 118mmHg on day 21 and raised to 141mmHg on day 51 (Figure 4C). Meanwhile, the extent of his fatty liver improved from severe on day 0 to moderate on day 51 (Table 5, Figure S1C) without any other treatment.

Case #1 was a 75-year-old Chinese woman diagnosed with hypertension. Before fasting, her BMI was 26.0 and her systolic blood pressure was 162mmHg. After fasting, her BMI was decreased to 22.5 on day 21 and 23.2 on day 51, while her systolic blood pressure declined to 138mmHg on day 21 and 204 on day 51 (Figure 4D). To our biggest surprise, she could still make her stools every 2-3 days even during fasting procedure.