Earlier studies have shown that the addition of vinegar in a carbohydrate-rich meal lowers glucose and insulin response in healthy individuals. The mechanism of how this is accomplished, however, remains unclear. The aim of this study is to examine the effect of vinegar on glucose and insulin response in patients with type II diabetes (T2D) in relation to the type of carbohydrates consumed in a meal.
Sixteen patients with T2D were divided into two groups, matched for age, gender and HbA1c. Patients in the first group (group A) were given a high-glycaemic index (GI) meal (mashed potatoes and low-fat milk) on two different days, with and without the addition of vinegar, respectively. In the second group (group B), patients were given an isocaloric meal with the same nutrient composition, but low GI (whole grain bread, lettuce and low-fat cheese). Postprandial plasma glucose and insulin values were measured every 30 min for 2 h.
In group A, the incremental area under the curve of glucose (GiAUC120) was lower after the addition of vinegar (181±78 mmol·min/l vs 311±124 mmol·min/l, P=0.04). The iAUC of insulin (IiAUC120) was also reduced, but the difference was of marginal statistical significance (2368±1061 μU·min/ml vs 3545±2586 μU·min/ml, P=0.056). In group B, the addition of vinegar did not affect either the GiAUC120 (229±38 mmol·min/l vs 238±25 mmol·min/l, P=0.56) or the IiAUC120 (2996±1302 μU·min/ml vs 3007±1255 μU·min/ml, P=0.98).
We conclude that the addition of vinegar reduces postprandial glycaemia in patients with T2D only when it is added to a high-GI meal.
Several studies have shown that plasma glucose levels measured 2 h after administering an oral glucose tolerance test can serve as an independent predictor of cardiovascular risk (Donahue et al., 1987; The DECODE Study Group, the European Diabetes Epidemiology Group, 1999). In addition, postprandial hyperglycaemia contributes to about 30–40% of the total daytime hyperglycaemia in patients with diabetes (Monnier et al., 2003). According to the guidelines issued by major scientific societies, correcting postprandial hyperglycaemia is part of the treatment strategy for the prevention of cardiovascular disease in patients with diabetes (AACE Diabetes Mellitus Clinical Practice Guidelines Task Force, 2007; American Diabetes Association, 2008). In addition, most authorities agree that both the amount of carbohydrates (CHO) as well as the type of CHO influence postprandial plasma glucose level, and the use of the glycaemic index (GI) and/or the glycaemic load (GL) can provide additional benefits over considering total CHO alone (Katsilambros et al., 2006).
Earlier studies have shown that the addition of vinegar in a CHO-rich meal lowers glucose and insulin responses in healthy individuals (Brighenti et al., 1995; Liljeberg and Björck, 1998; Johnston and Buller, 2005; Leeman et al., 2005; Ostman et al., 2005) and improves insulin sensitivity in individuals with insulin resistance and/or type II diabetes (T2D) (Johnston et al., 2004). It has also been shown that the addition of vinegar to a high-GL meal significantly reduces 60-min postprandial glycaemia in healthy individuals (Johnston and Buller, 2005). The mechanism, however, remains unclear and although it has been suggested that delayed gastric emptying might be responsible for the hypoglycaemic effect of vinegar, the results of relevant studies are conflicting (Brighenti et al., 1995; Liljeberg and Björck, 1998).
The aim of this study is to examine the effect of vinegar on postprandial glycaemic and insulin response in patients with T2D in relation to the GI of their meal.
Materials and methods
Sixteen patients with T2D, treated either with diet alone or with metformin monotherapy at a maximum dose of 850 mg daily, accepted to participate in the study. They were selected from a pool of patients with T2D attending the outpatient Diabetes Clinic of Laiko University Hospital in Athens, Greece. Participants were divided into two groups, matched for age, sex, body mass index, duration of diabetes and HbA1c. To be included in the study, patients needed to have an HbA1c <7.5%, a fasting plasma glucose <8.9 mmol/l and no history of major complications related to diabetes. All participants were given explanations about the purpose and the procedures of the study, which was approved by the ethics committee of the Laiko Hospital, and then signed an informed consent. The study required that patients visit the laboratory of the Diabetes Clinic for both their visits in the morning, after a 12-h fasting period.
Patients in the first group (group A) were given a high-GI mixed meal (meal A) on two different days, separated by 1 week (±2 days), with and without the addition of 20 g of wine vinegar (6% acetic acid, AB Vassilopoulos S.A.), respectively, corresponding to 1.2 g of acetic acid. In the second group (group B), patients were given an isocaloric meal with the same nutrient composition, but a low GI (meal B), using the same protocol (tested on two different days, with and without vinegar, respectively). Meal A was composed of pureed potatoes (50 g of mashed potato instant, GIOTIS S.A.) and low-fat (1%) milk (250 ml, FAGE S.A.), whereas meal B was composed of whole grain bread (100 g whole grain bread from wheat and rye, German Type, Katselis, NUTRIART S.A.), lettuce (55 g) and low-fat cheese (13% fat, 20 g, La Vache qui rit light, BEL S.A.). The macronutrient composition, GI and GL of the test meals are shown in Table 1. The GI and GL were calculated on the basis of validated international tables (Foster-Powell et al., 2002). Water (200 ml) was served with each meal. The meals were served in random order on the two occasions and were consumed steadily within 10 min.
Measurement of body weight, height, waist circumference and seated blood pressure were obtained during the two scheduled visits, pre-prandially. Vein blood samples were drawn before each meal and every 30 min for a total period of 2 h for analysis of plasma glucose and insulin levels. Plasma glucose was measured by the hexokinase method. Plasma insulin was measured in duplicate using BioSource INS-Irma immunoradiometric assay kits.
Analysis of the data was performed using the SPSS statistical package (SPSS 15.0, IL, USA). The incremental 2-h area under the curve (iAUC) for postprandial glucose and insulin plasma values were calculated for each patient test by using the trapezoidal rule and compared by using the paired samples, Student's t-test. In addition, the effect of vinegar on plasma glucose and insulin concentrations (at baseline, and 30′, 60′, 90′ and 120′ after the meal consumption) was assessed, in both groups, by using the one-way repeated-measures ANOVA. Between groups comparisons regarding demographic and clinical parameters have been performed by the independent samples Student's t-test. P-values <0.05 were considered statistically significant.
Table 2 shows that the two groups of participants were similar in terms of gender, age, body mass index, waist circumference, diabetes duration, HbA1c and proportion of patients treated with metformin. Fasting plasma glucose and plasma insulin values at baseline (time 0′) were also not statistically different between the two groups and between the two days of the experiment (within each group).
In group A (high-GI meal), the mean iAUC of glucose (GiAUC120) was significantly lower when vinegar was added than when the meal was consumed alone (181±78 mmol·min/l vs 311±124 mmol·min/l, P=0.04). Six out of eight participants of that group (A) showed a lower GiAUC120 when vinegar was added in the meal. Regarding plasma insulin, the mean iAUC of insulin (IiAUC120) was lower when vinegar was added than when the meal was consumed alone. However, this difference was of marginal statistical significance (2368±1061 μU·min/ml vs 3545±2585 μU·min/ml, P=0.056). Five out of eight participants in group A showed a lower iAUC120 after vinegar was added in the meal.
In group B, the addition of vinegar in the low-GI meal did not significantly affect either glucose or insulin response [GiAUC120: 229±38 mmol·min/l (with vinegar) vs 238±25 mmol·min/l (without vinegar), P=0.56; IiAUC120: 2996±1302 μU·min/ml (with vinegar) vs 3007±1255 μU·min/ml (without vinegar), P=0.98].
One-way repeated-measures ANOVA analysis (Figure 1) showed a significant lowering effect of vinegar addition on glucose response in group A (F=7.3, P=0.03), but not in group B (F=3.0, P=0.13). There was also a significant interaction effect between vinegar addition and glucose concentration over time (0–120 min, F=4.9, P=0.004) in group A, but not in group B (F=0.43, P=0.79). The interaction effect (in group A) was significant at times 60′ (P=0.034), 90′ (P=0.03) and 120′ (P=0.02) (Figure 1). The same type of analysis, when applied for insulin response (Figure 2), showed a marginal but non-significant lowering of plasma insulin concentration after vinegar addition in group A (F=5.2, P=0.054). In group B, vinegar addition did not affect insulin response (F=0.01, P=0.95). The interaction between vinegar addition in the meal and insulin concentration over time (0–120 min) was statistically significant in group A (F=3.7, P=0.03), but not in group B (F=0.84, P=0.46). The interaction effect (in group A) was significant at times 90′ (P=0.04) and 120′ (P=0.03) (Figure 2).
The main finding of this study is that in patients with T2D, postprandial hyperglycaemia is reduced when vinegar is added to a high-GI meal. However, postprandial hyperglycaemia is not affected by the addition of the same amount of vinegar to an isocaloric, of equal CHO-content, low-GI meal.
Earlier studies, in healthy adults, have shown that several organic acids (including propionic acid, lactic acid and acetic acid) exhibit a postprandial plasma glucose and insulin-lowering effect. The mechanism by which these substances produce this effect is not fully elucidated. In the case of acetic acid, some evidence suggests that the delay in gastric emptying (estimated by paracetamol measurements in the blood) may explain the reduced glycaemic and insulinaemic responses (Hunt and Knox, 1969; Liljeberg and Björck, 1998). Acidity and fatty acids are known to affect stomach motility through duodenal receptor-mediated mechanisms (Hunt and Knox, 1972; Lin et al., 1990). In another relevant study (Brighenti et al., 1995), however, gastric emptying as measured by ultrasonography was not affected by the addition of vinegar in a mixed meal. In addition to delayed gastric emptying, other possible implicated mechanisms include a lower rate of starch hydrolysis in the upper small intestine (Brighenti et al., 1995), suppression of disacharidase activity (Ogawa et al., 2000) and earlier induction of satiety (Ostman et al., 2005).
The effect of vinegar on postprandial glucose response in T2D has been examined earlier only in one study with 10 participating patients (Johnston et al., 2004). It was shown that the consumption of 20 g apple cider vinegar immediately before a CHO-rich mixed meal (87 g total CHO) improved (non-significantly) postprandial insulin and glucose fluxes and insulin sensitivity (estimated using a composite score). Postprandial increase in glucose after vinegar addition to a CHO-rich mixed meal has also been examined in type I diabetes (Mitrou et al., 2010), using artificial pancreas. It was shown that vinegar, compared with placebo, reduced glucose AUC0−240 by almost 20%. However, caution should be paid in patients with type I diabetes and gastroparesis, as it has been shown that in such patients, vinegar addition to food reduces the gastric emptying rate even further, and this might be a disadvantage regarding their glycaemic control (Hlebowicz et al., 2007).
In our study, the consumption of similar amount of vinegar together with a CHO-rich (51 g), high-GI (86) meal significantly reduced postprandial plasma glucose by 42% as indicated by GiAUC120. We were not able to compare our results with those of the study by Johnston et al. (2004), as they only reported incremental glucose values at 30 and 60 min. As it can be extrapolated from their published figure, however, at 60 min after glucose ingestion, in the vinegar group, plasma glucose seems to have decreased by 1 mmol/l, whereas in our study, it decreased by 2 mmol/l. Different characteristics of the two diabetic populations might be responsible for this difference (these are not provided in Johnston et al., 2004).
Earlier reports on healthy individuals have shown that when 20 g of vinegar were added to CHO-rich, high-GI mixed meals, glucose response decreased by 11–54% (Johnston and Buller, 2005: 54%; Liljeberg and Björck, 1998: 35%; Sugiyama et al., 2003: 25–35%; Brighenti et al., 1995: 30%; Ostman et al., 2005: 11%). The high decrease observed in our study might be attributed to the fact that patients with diabetes exhibit (by definition) higher postprandial glucose excursions than healthy persons, which leads to a higher proportional glucose-lowering effect of any anti-hyperglycamic intervention.
The novel finding of this study is that when the vinegar was added to an isocaloric mixed meal, with the same total-CHO content, but having a remarkably lower GI, the glucose-lowering effect of vinegar disappeared (Figure 1). Interestingly, an earlier study in healthy individuals (Johnston and Buller, 2005) showed that adding vinegar (in the form of apple cider vinegar, 20 g, 5% acetic acid) to a high-GI/high GL mixed meal (GI: 96, GL: 81) reduced the 60-min postprandial glycaemia significantly, by about 55%. The addition of the same amount of vinegar, however, on a mixed meal with similar GI (=91), but lower GL (=48), still reduced the 60-min glucose response, although not in a statistically significant proportion. The difference with this study is that while their meal exhibited a low-GL, but still a high-GI value (the difference in GL was due to a lower CHO content), our study had both low-GL and low-GI values.
In group B, the low-GI meal produced (as expected) a lower glycaemic and insulin response than the high-GI meal (without vinegar). It must be emphasized that these figures cannot be directly compared, as the two meals were tested in two different groups of diabetic patients (matched, however, for age, gender, body mass index, duration of diabetes and level of glycaemic control). The low GI of the meal was mainly due to its high-fibre content. Nevertheless, it was clearly shown that the addition of vinegar in the low-GI meal did not decrease postprandial glucose and insulin response (Figures 1 and 2).
To our knowledge, the effect of vinegar on high-fibre/low-GI meals has not been examined in the past. In the study by Johnston and Buller (2005), the GL of the low-GL mixed meal was 48, which is still quite higher than the one presented in our study (19.8, see Table 1), whereas the GI was 96 compared with only 38 in this study. The low-GI/GL meal in our study still produced a substantial increase in postprandial glucose and insulin levels, which was not affected by the addition of vinegar.
As mentioned above, the mechanism by which vinegar reduces postprandial glycaemia/insulinemia could be attributed to the decrease in the rate of gastric emptying, which in turn leads to slower CHO absorption. It can be hypothesized that the slow CHO absorption caused by a high-fibre/low-GI meal is not further affected by the effect of vinegar on gastric emptying. In other words, slow CHO absorption caused by fibre is not further delayed by vinegar. However, we believe that more studies are needed to elucidate the extent that other physiological mechanisms affect glucose absorption in the presence of acetic acid.
A limitation of this study is that the four different test meals (high/low GI, with/without vinegar) were tested in two different groups of patients with T2D, matched, however, for main demographic and clinical characteristics (Table 2). This approach was chosen to avoid the need to test every patient at four different days and confine the necessary experimental days to two per patient, as no patient was willing to participate having to be tested at four different occasions.
In conclusion, the addition of vinegar on high-GI meals might serve as a protective measure to avoid excessive postprandial rise in plasma glucose, in patients with ‘early’ T2D. Postprandial hyperglycaemia is generally recognized as an important factor affecting overall glycaemic control, whereas, according to several reports, is also believed to consist of an independent predictor of cardiovascular events (The DECODE Study Group, the European Diabetes Epidemiology Group, 1999). According to the results of this study, however, when the consumed meal had a low-GI value, vinegar addition did not further suppress postprandial glucose response.
We thank Mr Elias Kourpas for his assistance to write the article in proper English.