Disclosure: The authors declared no conflict of interest.
Article first published online: 26 MAR 2013
Copyright © 2012 The Obesity Society
Volume 21, Issue 2, pages 251–253, February 2013
How to Cite
Leader, N. J., Ryan, L., Molyneaux, L. and Yue, D. K. (2013), How best to use partial meal replacement in managing overweight or obese patients with poorly controlled type 2 diabetes. Obesity, 21: 251–253. doi: 10.1002/oby.20057
Relevant conflicts of interest/financial disclosures: Nothing to report. Full financial disclosures and author notes may be found in the online version of this article
- Issue published online: 26 MAR 2013
- Article first published online: 26 MAR 2013
- Accepted manuscript online: 3 OCT 2012 05:37PM EST
- Manuscript Accepted: 1 AUG 2012
- Manuscript Received: 29 MAR 2012
To compare patient compliance and benefits, over 12 months, of 1 versus 2 partial meal replacement (PMR) for the management of overweight/obese subjects with inadequately controlled type 2 diabetes.
Design and Methods:
Thirty-six overweight patients with inadequately controlled type 2 diabetes (BMI > 27 kg/m2 and HbA1c > 7.5% [58 mmol/mol]) were randomized to receive 1 or 2 PMR/day, while maintaining usual lifestyle. Subjects were seen monthly and adjustment of medications was made to prevent hypoglycemia. Compliance was assessed by counting unused sachets.
Patients on 2 PMR/day lost almost 4 kg compared with only 0.5 kg in the 1 PMR/day group. This difference was statistically significant (P < 0.05). Overall PMR was about 30% as effective as in our previous study on total meal replacement. Reductions in weight, waist, and HbA1c were better in the 2 PMR/day group while patient dropout and compliance were not worse over a 12-month period.
PMR provides a further management option for overweight/obese individuals with type 2 diabetes. The initial recommendation should be 2 PMR/day.
The treatment of type 2 diabetes in overweight/obese patients is often suboptimal. Lifestyle changes are difficult to achieve and increase in medication commonly leads to weight gain with little improvement in glycemic control. Very low calorie diet as total meal replacement (TMR) undoubtedly improves weight loss and glycemic control (1) but is difficult to maintain. Partial meal replacement (PMR) as part of an overall treatment strategy has been shown to be beneficial (2-4) and to an extent proportional to the number of meal replacements used over a period of several years (5). What is not known is whether advising patients to have PMR for one or two meals is more effective in the longer term. In a “tortoise and hare” analogy, a 1 PMR/day may be initially less effective but more sustainable in the longer term. This was tested prospectively over a period of 12 months.
Methods and Procedures
Thirty-six patients (15 male) with type 2 diabetes (BMI 39.3 ± 5.3 kg/m2, HbA1c 9.4 ± 1.2% [79 ± 13 mmol/mol]) were randomized to receive 1 or 2 PMR/day using Optifast®. A “taste test” was given before randomization and subjects only proceeded if they confirmed a willingness to remain on it for 12 months. Patients were to maintain one daily “most important social meal” but not to compensate calorically at the others. All patients had previously received standard diet advice and only minimal further lifestyle/diet advice was provided. Subjects were assessed monthly and diabetic medications adjusted to avoid hypoglycemia, based on patient reporting of appropriate symptoms or documented episode of blood glucose level < 4.0 mmol/l. Patients were given sufficient meal replacement sachets at no charge and compliance was assessed by counting the number of leftover sachets each month. If a patient failed to attend a scheduled appointment, they were contacted. If they did not attend a rebooked appointment or were unable to be contacted or chose to leave the study, they were deemed to be drop-outs.
Data was compared using the NCSS 2004 statistical package according to Intention to Treat randomization, with the last available results carried forward, or as the Completers Cohort at each time point. Continuous data was presented as mean or median and compared with Two Sample or Mann-Whitney test. Paired t or Wilcoxon Signed-Rank test were used to assess the treatment over time. Chi-square was used to compare groups. A P value of P < 0.05 was considered statistically significant.
Informed consent was given by patients and the study was approved by the Ethics Committee of the Institution.
The clinical profiles of the two groups were comparable at baseline. Those in the 1 PMR/day group had an average age of 56.7 (±8.9) years and a median duration of diabetes of 10 (5-12) years. In the 2 PMR/day group, the corresponding values were 55.3 (±8.7) years and 8 (4-14) years. Many of the subjects were on two oral anti-diabetic agents (70% in 1 PMR/day group and 44% in the 2 PMR/day group). Seventy percent of the 1 PMR/day group and 65% of the 2 PMR/day group were also on insulin, median of 112 (77-155) and 124 (74-156) units per day, respectively. At the end of the study, neither group had a significant reduction in the number of oral anti-diabetic agents they were taking. However, the 1 PMR/day group was taking 7.4 ± 10.7 units less of insulin/day and for the 2 PMR group 12.1 ± 19.6 units less/day.
Regardless of their grouping, those that completed the first 3-month period had a significant reduction in waist circumference, weight, and HbA1c when compared with baseline (Table 1). The benefits were more pronounced and better retained throughout the follow up period in the 2 PMR/day group which at the end of the study had an HbA1c 0.9% lower and a weight loss of 4 kg compared with the baseline. The benefits became progressively smaller in the 1 PMR/day group during the 12 months follow up. Dropout rate and compliance in taking the PMR were not statistically different between groups.
|Parameters||1 PMR/day||2 PMR/day|
|Baseline and Δ waist (cm)||120.9 ± 12.9||125.3 ± 14.5|
|3 months||−4.3 ± 3.1b||−4.5 ± 3.1b||−7.4 ± 2.8b||−6.9 ± 3.7b|
|6 months||−4.4 ± 5.2b||−4.8 ± 4.7b||−8.8 ± 3.8b||−7.4 ± 3.7b|
|9 months||−4.3 ± 4.8b||−5.2 ± 4.6b||−7.0 ± 8.7||−6.7 ± 2.9b|
|12 months||−1.1 ± 6.2||−3.7 ± 5.4b||−11.3 ± 6.4b||−7.9 ± 5.2b|
|Baseline and Δ weight (kg)||103.3 ± 19.7||111.4 ± 20.8|
|3 months||−1.8 ± 2.9b||−1.8 ± 2.7b||−4.1 ± 4.5b||−4.0 ± 4.2b|
|6 months||−1.7 ± 3.7||−1.9 ± 3.3b||−4.8 ± 5.8b||−3.7 ± 4.9b|
|9 months||−2.7 ± 2.8b||−1.8 ± 3.2b||−5.4 ± 5.5b||−4.1 ± 4.5b|
|12 months||0.5 ± 5.1||−0.5 ± 4.2||−5.9 ± 6.2||−3.8 ± 4.8b|
|Baseline and Δ HbA1c (%)||9.3 ± 1.3 (78 ± 15 mmol/mol)||9.5 ± 1.1 (80 ± 12 mmol/mol)|
|3 months||−0.6 ± 0.9b (−6 ± 9 mmol/mol)||−0.5 ± 0.9b (−6 ± 10 mmol/mol)||−1.2 ± 1.0b (−13 ± 12 mmol/mol)||−1.0 ± 1.1b (−11 ± 12 mmol/mol)|
|6 months||−0.6 ± 1.1b (−7 ± 12 mmol/mol)||−0.6 ± 1.1b (−6 ± 12 mmol/mol)||−1.0 ± 1.4 (−11 ± 15 mmol/mol)||−0.8 ± 1.0b (−9 ± 11 mmol/mol)|
|9 months||−0.3 ± 1.6 (−5 ± 17 mmol/mol)||−0.4 ± 1.3 (−5 ± 14 mmol/mol)||−1.2 ± 1.4 (−13 ± 15 mmol/mol)||−1.0 ± 1.1b (−10 ± 12 mmol/mol)|
|12 months||−0.2 ± 1.2 (−3 ± 13 mmol/mol)||−0.4 ± 1.0 (−4 ± 11 mmol/mol)||−0.6 ± 1.4 (−10 ± 16 mmol/mol)||−0.9 ± 1.3b (−9 ± 14 mmol/mol)|
|Dropouts (cumulative %)||3 months||15||19|
|Compliance (%)||3 months||83 ± 19||77 ± 25|
|6 months||82 ± 30||73 ± 26|
|9 months||79 ± 27||73 ± 32|
|12 months||92 ± 14||83 ± 19|
The management of overweight/obese patients with diabetes is particularly difficult due to the inability to modify lifestyle in the long term and because pharmacological treatment often promotes weight gain, thus creating a vicious cycle. While the effectiveness of TMR is established, this intensity of treatment is usually only sustainable for a few months and benefits gradually wane. In the Look AHEAD study, PMR was a component of the overall intensive intervention for treatment of type 2 diabetes. There was found to be a good dosage response relationship between weight loss and the extent of PMR used, as determined by patients' choice (5). Naturally, more extensive PMR would be expected to bring more benefit. However, a greater degree of PMR may also be associated with higher patient dropout and noncompliance. Therefore, we have compared a one or two meal PMR to determine which has the better compliance, metabolic response, and sustainability.
PMR by either strategy was found to reduce waist circumference, weight, and hyperglycemia. Not surprisingly, it is only about 30% as effective as TMR as previously reported by our group (1). The current study showed that 2 PMR/day always produced better outcomes for each of these parameters at all time points studied, up to 12 months of follow up. This trend is evident whether assessed by intention to treat analysis or by only examining the completers at each time point. The superior results of 2 PMR/day occurred without any significant detriment to compliance or dropout rate. Therefore, results of this study suggest that, for clinical use, 2 PMR/day is a better recommendation. It is possible that with patients experiencing better results initially, they are more likely to continue with a more vigorous treatment regimen.
Several factors that can affect our results need to be acknowledged. The relatively small number of subjects studied reduced the power of the study. The duration of the study was only 12 months, although this would be considered reasonably long for dietary intervention. Changes in weight and glycemia would be confounded by the necessity and the extent of down titrating dosage of medications to avoid hypoglycemia. Patients were on a variety of oral agents and insulin regimen, making it difficult to standardize a titration protocol with the number of patients studied. Adjustment of medications was, therefore, only made on the basis of clinical symptoms of hypoglycemia or a recorded blood glucose level of <4.0 mmol/l. The baseline treatment and the intensity of down titration in medications could obviously affect metabolic response and weight changes. An over zealous reduction in dosage of medications, particularly insulin, would produce better weight reduction but at the expense of less improvement in glycemic control, and vice versa. While weight and HbA1c changes are clearly important outcome measures, in the context of this study, we are more focused on comparing the compliance and dropout rate in response to the 1 or 2 meal PMR/day in a reasonably real clinical setting. As mentioned above, the results of this study have established that 2 PMR/day seems to be better without compromising patient acceptance.
It is also noteworthy that there were differences in our patients' knowledge regarding diet, weight loss, and glycemic control despite all patients having been previously educated in these aspects. There were significant variations in their behavioral aspects of eating but again we chose not to intervene so as to test specifically the effects of PMR alone and in a setting more similar to the real world situation. We, therefore, believe that our findings have relevance to everyday clinical care.
- 2Why WAIT program: a novel model for diabetes weight management in routine clinical practice. Obes Manage 2008; 4: 176-184., , , et al.