Impact of bariatric surgery on type 2 diabetes in morbidly obese patients and its correlation with pre-operative prediction scores.

BACKGROUND: Bariatric surgery, besides causing significant weight reduction, leads to improvement in type 2 diabetes mellitus (T2DM). However, there is a scarcity of data on the prediction of diabetes resolution in non-Western population.
OBJECTIVE: To evaluate the impact of bariatric surgery on T2DM and to assess the accuracy of pre-operative scoring systems in predicting remission. STUDY
SETTING: A tertiary care academic centre, India.
METHODOLOGY: We used a retrospective cohort of all diabetic patients (n = 244) who underwent bariatric surgery at our centre in the past 10 years. The cohort was followed up for diabetes remission, and pre-operative scoring systems were analysed against the observed results.
RESULTS: Of 244 patients, we were able to contact 156 patients. The median period of follow-up was 38 months. The mean body mass index (BMI) of the study group decreased from 45.4 to 33.4 kg/m2 (%excess BMI loss = 61.2%). The number of patients dependent on oral anti-diabetic pharmacotherapy and on insulin decreased from 133 (85.3%) to 40 (25.6%) and from 31 (19.9%) to 7 (4.5%), respectively. Remission was analysed for 96 patients, who submitted complete biochemical investigations. The median follow-up period for this sub-cohort was 36 months. 38 (39.6%) patients were in complete remission, 15 (15.6%) patients in partial remission and 34 (38.5%) patients showed an improved glycaemic control. The three pre-operative scores, Advanced-DiaRem, DiaRem and ABCD, showed predictive accuracies of 81.1%, 75.6% and 77.8%, respectively.
CONCLUSIONS: Besides leading to excess BMI loss of 61.2%, bariatric surgery also resulted in diabetes remission in 55.2% of the patients. Amongst various pre-operative scores, Advanced-DiaRem has the highest predictive accuracy for T2DM remission.

INTRODUCTION

The risk of developing type 2 diabetes mellitus (T2DM) is 50-fold higher for the obese population, as compared to their non-obese counterparts.[12] As a result, the recent increase in the prevalence of obesity has caused a parallel rise in the prevalence of diabetes. Bariatric surgery causes significant weight reduction and has also shown to cause improvement in glycaemic control of diabetic patients.[34] Remission rates are as high as 75%–80% at 1 year of follow-up[5] and then decrease progressively with time. Medium-term and long-term remission rates have been reported to be at around 60%–65% and 50%, respectively.[678] The success of metabolic surgery in causing remission of T2DM depends on various factors such as body mass index (BMI), age, duration of diabetes and pre-operative glycaemic control. Various scoring systems such as ABCD,[9] DiaRem[10] and Advanced-DiaRem (Ad-DiaRem)[11] derived from these variables Appendix 1 have been developed to predict the probability of diabetes remission after bariatric surgery.

The DiaRem and Ad-DiaRem scoring systems were developed through studies on the Western population. A study done in 2017 targeted a more diversified cohort and suggested that the predictive value of DiaRem score may vary in different ethnic groups.[12] Asians are known to have a higher prevalence of T2DM than their Western counterparts. They also might not have the same response to bariatric surgery as the Western population as evidenced in a meta-analysis study.[13] Thus, there is a need to study these scores, especially Ad-DiaRem which has not been evaluated in the non-Western population. ABCD and DiaRem scores also need to be studied in the Indian population.

We report on follow-up of diabetic patients who underwent bariatric surgery and on the correlation of existing pre-operative predictive scores with the actual percentage of remission obtained.

METHODOLOGY

A prospectively maintained database of all the patients who underwent bariatric surgery at our tertiary care teaching hospital in the past 10 years (February 2008 to February 2018, n = 769) was analysed retrospectively after appropriate ethical clearance. Patients had undergone one of the three procedures, laparoscopic sleeve gastrectomy (LSG) or Roux-en-Y gastric bypass (RYGB) or one-anastomosis gastric bypass (OAGB). Patients were counselled regarding all there procedures. Patients with long duration of T2DM (>5 years) or requiring insulin or diabetes-related end-organ damage or severe reflux symptoms were advised to undergo RYGB/OAGB. The final decision was left with the patient. All patients were maintained under regular endocrinology follow-up, before the consideration of bariatric surgery and also after the operation. At our institution, we generally advise the patients for post-operative follow-up 3 monthly for the 1st year, 6 monthly for the 2nd and then yearly. In this population, 244 (31.7%) patients suffered from T2DM. The pre-operative data collected included date of operation; contact details of the patient, sex, age and BMI at the time of registration; duration of diabetes; use of anti-diabetic medications (oral/insulin) and anti-hypertensive medications (if any); type of bariatric procedure and the pre-operative blood investigations. Pre-operative blood investigations included fasting glucose levels, glycosylated haemoglobin (A1C), C-peptide and the lipid profile of the patient. The follow-up methodology included an initial phone call to inform the patients and ask about reduction in weight and use of medications. For further investigations, patients were advised to visit our centre or get the investigations done and send the reports via mail.

The weight loss has been reported as percentage excess BMI loss (PEBMIL), calculated using the formula: %EBMIL = (ΔBMI/[Initial BMI – 25]) × 100%[14] as well as total weight loss (TWL).

Remission of type 2 diabetes mellitus

Preoperatively, patients were considered to have T2DM if they had A1C >6.5 or FBS >126 mg/dl. Complete remission of T2DM was considered for A1C <6% and fasting glucose <100 mg/dl. For partial remission, biochemical cut-offs were 6.5% for A1C and 126 mg/dl for fasting glucose levels. For both complete and partial remissions, the patient was required to be off anti-diabetic medications at the time of follow-up. The risk ratio was calculated to study the effect of various pre-operative factors upon remission of T2DM. In addition, all patients were scored according to ABCD, DiaRem and Advanced-DiaRem scores using their available pre-operative data. The scores were evaluated against actual results of remission in our cohort, to study the predictive value of these scoring systems for the Indian population. The cut-offs for predicting remission by Ad-DiaRem (remission at scores <10) and DiaRem scores (remission at scores <7) were taken from existing literature.[1011] The cut-off for predicting remission by ABCD score was found using Youden index on our study results. A higher value on ABCD score is associated with a greater probability of remission, while a lower value on DiaRem and Ad-DiaRem is more favourable.

Statistical analysis

All the data were analysed using STATA Ver. 14.0. Continuous variables were described as means and standard deviations and were analysed using various parametric and non-parametric tests. Duration of T2DM was stated in terms of the median value. Categorical data were compared using Fisher’s exact and Pearson’s Chi-squared tests. The efficacy of various pre-operative predictive scores was compared using logistic regression analysis and open-source statistical software OpenEpi. P < 0.05 was considered statistically significant.

RESULTS

Of the 244 patients, 156 were contacted via phone call and 96 met to consultation.

Entire cohort (n = 156)

Starting from April 2018, we were able to contact 156 patients of the 244 diabetic patients who had undergone bariatric surgery (64%, n = 156). 64.1% of our study population were female. Out of 156 patients, 44 (28.2%) patients had undergone RYGB, 97 (62.2%) had undergone LSG and 15 (9.6%) had undergone OAGB. The median period of follow-up was 38 months (interquartile range [IQR]: 13–59.5). The mean BMI of patients decreased from 45.4 ± 8.0 kg/m2 preoperatively to 33.4 ± 6.3 kg/m2 at the time of follow-up (PEBMIL = 61.2% ± 22.6% and %total body weight loss = 25.6% ± 9.5%).

Bariatric surgery caused a large reduction in the number of patients on oral anti-diabetic pharmacotherapy, from 133 (85.3%) patients to 40 (25.6%) patients in 2018 (P < 0.01). Similarly, the number of patients dependent on insulin also decreased from 31 (19.9%) to 7 (4.5%) (P < 0.05).

Changes in glycaemic control (n = 96)

By July 2018, follow-up biochemical investigations were received for 96 patients of the 244 diabetic patients who underwent bariatric surgery at our centre. Out of these 96 patients, 54 had undergone LSG, 35 had undergone RYGB and 7 had undergone OAGB procedure. The median duration of T2DM before surgery was 36 months (IQR: 12–111). The mean age of these patients was 45.2 ± 9.7 years. 68.8% of them were female. The median duration of follow-up was 31.5 months (IQR: 9.25–50.75). BMI of the sub-cohort decreased from 45.5 ± 8.0 kg/m2 preoperatively to 33.9 ± 6.8 kg/m2 at the time of follow-up (P < 0.05). The mean fasting glucose levels and A1C of this sub-cohort had also decreased from 135.7 mg/dl to 106.1 mg/dl (P < 0.05) and from 7.9% to 6.4% (P < 0.05), respectively.

Analysis of remission

Out of 96 patients, 38 (39.6%) achieved complete remission and 15 (15.6%) achieved partial remission. 34 (38.5%) patients did not meet the biochemical cut-offs for remission, but showed an improvement in glycaemic control, relative to their baseline levels. For further analysis, the patients were divided into remitters and non-remitters, and their pre-operative and follow-up data were compared [Table 1, pre-operative data and Table 2, follow-up data].

Table 1

Pre-operative data according to remitters versus non-remitters (n=96)

	Whole cohort (%)	Remission (%)	No remission (%)	P
Total number of patients	96	53 (55.2)	43 (44.8)	-
Female, n (%)	64 (67.4)	40 (75.5)	24 (57.1)	0.06
Age (years)	45±9.6	44.1±10	46.2±9	0.30
BMI (kg/m2)	45.4±7.9	46±8.5	44.6±7.0	0.38
Duration of T2D (median, months)	36	18	90	<0.01*
Use of insulin	17 (18.3)	2 (3.8)	15 (37.5)	<0.01*
Patients on diabetic pharmacotherapy	88 (93.6)	48 (90.6)	40 (97.6)	0.17
Number of glucose-lowering agents used by patients	2±1.1	1.7±1.1	2.3±0.9	<0.01*
FBS (mg/dl)	135.7±52.2	132.3±49.4	140.1±56.4	0.47
A1C (%)	7.9±1.7	7.4±1.7	8.5±1.5	<0.01*
LSG	54	29 (53.7)	25 (46.3)	-
RYGB	35	18 (51.4)	17 (48.6)	-
OAGB	7	6 (85.7)	1 (14.3)	-

*P<0.05 indicates significant result. T2D: Type 2 diabetes, BMI: Body mass index, FBS: Fasting blood sugar, LSG: Laparoscopic sleeve gastrectomy, RYGB: Roux-en-Y bypass, OAGB: One-anastomosis gastric bypass, A1C: Glycosylated haemoglobin. Number of diabetic patients considered=244, Number of patients with biochemical investigations at the time of follow-up=96

Table 2

Follow-up data according to remitters versus non-remitters (=96)

	Whole cohort	Remission	No remission	P
Total number of patients (%)	96	53 (55.2)	43 (44.8)	-
BMI (kg/m2)	34.2±8	33.4±7	35.1±9.1	0.33
%EBMIL	68±56.5	74.1±63.8	60.2±45.4	0.24
FBS (mg/dl)	106.1±30.4	89.4±10.2	126.1±30.3	<0.01*
A1C (%)	6.4±1.4	5.6±0.5	7.4±1.4	<0.01*

*P<0.05 indicates significant result. BMI: Body mass index, %EBIL: Percentage excess BMI loss, FBS: Fasting blood sugar, A1C: Glycosylated haemoglobin. Number of diabetic patients considered=244. Number of patients with biochemical investigations at the time of follow-up: 96

The duration of T2DM before seeking bariatric surgery showed a significant difference between remitters and non-remitters (P < 0.01; median values: 18 months against 90 months). 15 patients (37.5%) amongst non-remitters used insulin preoperatively, as compared to 2 patients (3.8%) amongst remitters; P < 0.01). Furthermore, though fasting glucose levels did not show much difference, pre-operative A1C was significantly lower for the group that showed remission. These patients also had significantly higher mean pre-operative c-peptide levels than those who did not (4 ng/dl against 2.6 ng/dl, P < 0.01).

Patients who showed remission had a greater PEBMIL that those who did not. At the time of follow-up, the average fasting glucose levels and A1C levels were significantly different between the two groups. Pre-operative factors including insulin use, duration of diabetes >5 years, use of >2 glucose-lowering agents, A1C levels >6.5 and c-peptide levels >2 ng/ml, BMI >50 Kg/m2 and male gender were studied. The risk ratios obtained, along with their confidence intervals, are summarized in Table 3.

Table 3

Impact of pre-operative factors on diabetic remission

Pre-operative factor	RR	95% CI	P
Male sex	0.78	0.51-1.21	0.284
Use of insulin	0.165	0.04-0.62	0.007
Use of>2 glucose-lowering agents	0.60	0.43-0.86	0.004
A1C>6.5	0.67	0.45-0.93	0.017
C-peptide>2 ng/ml	1.67	1.15-2.41	0.007
BMI>50	1.23	0.85-1.78	0.275
Duration of T2DM>5 years	0.35	0.19-0.64	0.0005

CI: Confidence interval, BMI: Body mass index, FBS: Fasting blood sugar, T2DM: Type 2 diabetes mellitus, A1C: Glycosylated haemoglobin, RR: Risk ratio

Scoring of patients according to ABCD, DiaRem and Advanced-DiaRem scores

All patients who were analysed for remission of T2DM were scored according to three existing predictive scores: ABCD, DiaRem and Advanced-DiaRem. Due to lack of complete pre-operative investigations, 54 patients could be scored according to ABCD scores, while 90 patients were scored according to DiaRem and Ad-DiaRem scores. The results from the three scoring systems are compared in Table 4.

Table 4

Relevance of different pre-operative scores in predicting diabetic remission

	ABCD score	DiaRem score	Advanced- DiaRem score
Number of patients scored	54	90	90
Range of score	0-10	0-22	0-21
Cut-off for remission	>6	<7	<10
Discriminative capacity (area under ROC curve)	80.7	83.7	84.3
Sensitivity	76.5	65.4	84.6
Specificity	80.0	89.5	76.3
PPV	86.7	89.5	83.0
NPV	66.7	65.4	78.4
Predictive accuracy	77.8	75.6	81.1

ROC: Receiver operating characteristics, PPV: Positive predictive value, NPV: Negative predictive value

Analysis according to the type of procedure

Patients of the three procedures did not vary significantly in their pre-operative biochemical investigations and use of anti-diabetic pharmacotherapy. Patients of OAGB had a higher mean BMI at registration (49.9 ± 5.1 kg/m2) than patients of RYGB (43.6 ± 5.2 kg/m2, P = 0.023) or patients of LSG (45.5 ± 9.1 kg/m2, P = 0.14).

At follow-up, patients of all three procedural arms showed a decrease in use of insulin and anti-diabetic pharmacotherapy. Further analysis revealed that LSG and RYGB did not show any significant difference in their efficacy of lowering anti-diabetic pharmacotherapy (oral/insulin), c-peptide levels or on the biochemical profile (fasting glucose and A1C) of the patients. The median pre-operative duration of T2DM in the patients undergoing LSG, RYGB and OAGB was 30, 60 and 36 months, respectively. The details are shown in Table 5.

Table 5

Comparison of different procedures in their efficacy in improving glycaemic control

Parameter	Procedure	At the time of registration	At the time of follow-up
n=156 (%)
Oral pharmacotherapy	LSG	83.5	24.7
	RYGB	90.9	36.4
	OAGB	86.7	0
Insulin use	LSG	15.5	5.2
	RYGB	29.6	4.6
	OAGB	20	0
n=96
BMI (kg/m2)	LSG	45.7±9.3	34.7±7.7
	RYGB	44.5±6.0	32.7±5.4
	OAGB	49.1±5.0	33.8±6.4
Fasting glucose (mg/dl)	LSG	132.9±54.8	108.6±31.6
	RYGB	141.5±54.0	106.8±29.0
	OAGB	130.7±27.8	85.1±18.7
A1C (%)	LSG	7.7±1.5	6.5±1.5
	RYGB	8.2±2.1	6.4±1.3
	OAGB	7.3±1.0	6.0±1.2

For each of these investigations, P value between follow-up results of LSG and RYGB was not significant. BMI: Body mass index, LSG: Laparoscopic sleeve gastrectomy, RYGB: Roux-en-Y bypass, OAGB: One-anastomosis gastric bypass, A1C: Glycosylated haemoglobin. Number of diabetic patients considered=244. Number of patients who could be contacted at the time of follow-up=156. LSG: 97 patients, RYGB: 44 patients, OAGB: 15 patients. Number of patients with biochemical investigations at the time of follow-up=96. LSG: 54 patients, RYGB: 35 patients, OAGB: 7 patients

DISCUSSION

We report an overall remission rate of 55.2% (complete + partial), after a median follow-up period of 36 months. Another 38.5% of patients showed an improvement in glycaemic control. We compared our results to existing studies that analyse the outcomes of bariatric surgery in diabetic patients, after similar follow-up periods. The 3-year follow-up analysis of the STAMPEDE trial on patients of RYGB and sleeve gastrectomy showed that at follow-up, 47.4% of patients had A1C <6.5% (i.e., in complete/partial remission). The lower remission rate reported here might be attributed to the worse pre-operative glycaemic control of their study cohort (A1C at 9.4%).[3] It is noteworthy that the outcome measure in STAMPEDE trial was based solely on biochemical cut-offs for glycaemic control, regardless of the patient’s use of pharmacotherapy. In our study, 10 (10.5%) patients had A1C and fasting glucose levels below the required biochemical cut-offs for diabetes remission, but were on oral anti-diabetic pharmacotherapy at the time of follow-up and hence were considered under improved diabetic status and not under remission. A recent study from India by Nasta et al.[15] reported remission in 82.7% patients at 3 years based on discontinuation of anti-diabetic medication without regard to biochemical parameters. In comparison, we used stricter criteria based on biochemical investigations and hence obtained a lower rate of remission. Another recent study involving the Indian subcontinent (Misra et al.)[16] analysed 42 T2D patients for outcomes following LSG; and reported a diabetic remission rate (with A1C <6.5%) of 71.4% at the end of 3 years. However, the cohort studied by Misra had a mean fasting glucose level of 115.3 mg/dl and mean A1C level of 6.8%, in comparison to the fasting glucose level of 135.7 mg/dl and A1C of 7.9% of our group. This suggests a better pre-operative glycemic control in their study cohort, which might have contributed to the higher remission rate observed by them.

Analysis of pre-operative factors

In our study, the mean BMI decreased by 26.4% of original level. This is in agreement with other studies of similar follow-up durations, that have shown similar decrease of BMI by 28-30% of the original levels.[615] We also observed a significant reduction in dependence on anti-diabetic pharmacotherapy, the use of anti-hypertensive medications and a decrease in the mean c-peptide level. A decrease in C-peptide levels indicates a decrease in insulin secretion, which suggests a decrease in insulin resistance.[1718]

Results from our data suggest that pre-operative factors like shorter duration of T2DM before surgical intervention, lower values of A1C (i. e. better pre-operative glycemic control), a higher C-peptide level and lack of insulin pharmacotherapy favour remission in diabetic patients. A recent study (Jindal et al.)[19] analysed diabetic remission after LSG in 104 T2D patients and reported similar results. Jindal et al. observed that the beneficial effect of LSG is more in younger patients with better pre-operative glycaemic control and shorter duration of disease before surgical intervention. In addition, they also noted a positive correlation between pre-operative BMI values and T2D remission rates. The favourable association between shorter duration of T2DM and diabetes remission supports the International Diabetes Federation position of considering bariatric surgery early in the treatment algorithm for morbidly obese diabetic patients.[20] We also noted that remitters had a higher value of pre-operative c-peptide level than non-remitters. A higher c-peptide value suggests that the pancreas has not yet undergone severe damage, and surgical intervention will have a higher chance of causing diabetes remission.[21] Various studies have shown superiority of bariatric surgery over intensive medical management in morbidly obese patients. However, all morbidly obese diabetic patients do not achieve remission after surgery. There is also a need for regular monitoring and nutritional supplementation post-surgery. The patients should be carefully selected based on their chances of remission. The various factors such as age, duration of diabetes, C-peptide level, baseline BMI and visceral fat have been combined into scores (ABCD, DiaRem and Advanced-DiaRem) to predict the remission of T2DM following bariatric surgery. Ad-DiaRem score showed the highest predictive accuracy of 81.1%. This is probably because of its more intensive criteria, which, unlike ABCD and DiaRem scores, take into account both the glycaemic control of the patient and duration of diabetes before surgical intervention. DiaRem score does not consider the pre-operative duration of diabetes when calculating the probability of remission, while ABCD score does not consider the pre-operative glycaemic control of patients. Contrary to this, statistical analysis on our data has shown that both shorter duration of diabetes and better pre-operative glycaemic control have a significant favourable effect on probability of achieving diabetes remission.

It is noteworthy that none of the three scores were originally made for prediction of diabetes remission beyond short-term follow-up period. Advanced-DiaRem and DiaRem were both formulated for predicting short-term success rates of RYGB procedures in diabetic patients of the Western population. Our study was an external validation of these scores and showed that they can also be used for predicting diabetes remission after longer periods of follow-up in LSG and RYGB patients of non-Western population.

Evaluation of results according to the type of procedure

The PEBMIL was nearly equal between patients of sleeve gastrectomy and RYGB (66.3% for OAGB vs. 60.9% for LSG and 59.6% for RYGB procedures). LSG and RYGB did not show significant differences in their efficacy in improving the biochemical profile and use of anti-diabetic pharmacotherapy in diabetic patients. While there is a 54.7% remission rate observed for LSG patients, there is a similar remission rate of 51.4% in the group of patients who underwent RYGB. This was contrary to the general consensus that due to the added component of intestinal shortening and malabsorption, intestinal bypass procedures (such as Roux-en-Y gastric bypass) are associated with greater rates of diabetes remission.[22] The similar remission rates obtained might be attributed to worse pre-operative glycaemic control (higher mean A1C and mean fasting glucose level) and a longer duration of T2DM in patients who underwent RYGB procedure. The compliance of patients in regard to the anti-diabetic medications was not studied which could have been a confounding factor in the procedure-wise analysis.

At present, no standard guidelines exist that would dictate the choice of bariatric procedure. Some recent studies have shown good remission rates after sleeve gastrectomy, comparable to remission rates after RYGB procedures. A retrospective study conducted in 2013 in France compared different bariatric procedures in their efficacy in resolving type 2 diabetes. The study showed a non-significant difference in remission rates between LSG (62.5%) and RYGB (52%) procedural arms. The pre-operative A1C was higher for RYGB patients than for sleeve gastrectomy and hence RYGB patients had a relatively poor outcome.[23] Several studies have shown an enhancement in levels of insulinotropics such as GLP-1(Glucagon-LikePeptide-1) after both procedures.[724] It may be argued that though being considered as a majorly restrictive procedure, sleeve gastrectomy also has metabolic effects that could have resulted in remission rates being comparable to those after RYGB procedures.

The OAGB arm (one-anastomosis gastric bypass) showed a greater excess BMI loss and better glycaemic control in patients than both RYGB and LSG. At follow-up, the mean excess BMI loss observed was 66.3% for OAGB, 60.9% for LSG and 59.6% for RYGB procedures. The higher degree of improvement in glycaemic control in OAGB patients might have resulted from the greater malabsorptive component of the procedure. However, the study cohort included only 7 patients in the OAGB arm and, hence, was not powered enough to yield any significant results.

All bariatric surgery patients were operated according to standard procedure guidelines and by a single surgeon. However, there were certain limitations to our study. It was a retrospective, single-centre study and a higher follow-up would have been better. Another important limitation was the lack of pre-operative C-peptide levels for a significant part of the cohort, because of which ABCD score could only be assessed on 54 patients out of 96.

CONCLUSIONS

Bariatric Surgery has an excellent impact on T2M in Indian patients and leads to remission/improvement in majority (94%) of patients. Pre-operative predictive scores, namely, ABCD score, Advanced-DiaRem and DiaRem, can be applied in the non-Western setting also.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

ABCD scoring system[9]

Factor	Score
Age (years)
<40	1
>40	0
BMI
<27	0
27-34.9	1
35-41.9	2
>42	3
C-peptide (ng/ml)
<2	0
2-2.9	1
3-4.9	2
>5	3
Duration of DM (years)
>8	0
4-8	1
1-3.9	2
<1	3
Total score	0-10

Ad-DiaRem (Advanced- DiaRem) scoring system[11]

Factor	Score
Age (years)
15-41	0
42-52	3
53-69	5
HbA1C (%)
4.5-6.9	0
7.0-7.4	2
7.5-18.4	4
Treatment with insulin
No	0
Yes	3
Other glucose lowering agents*
No	0
Yes	1
Number of glucose lowering agents**
0	0
1	1
2	2
>3	3
Duration of DM (years)
0-6.9	0
7.0-13.9	3
>14	5
Total score	0-21

ISA: Insulin-sensitizing agent. *includes sulfonylurea and ISA other than metformin, **includes all glucose-lowering agents (including insulin)

DiaRem (Diabetes Remission) scoring system[10]

Factor	Score
Age (years)
<40	0
40-49	1
50-59	2
>60	3
HbA1C (%)
<6.5	0
6.5-6.9	2
7.0-8.9	4
>9.0	6
Other diabetic drugs
No sulfonylurea or ISA other than metformin	0
Sulfonylurea or ISA other than metformin	3
Treatment with Insulin
No	0
Yes	10
Total score	0-22

ISA: Insulin-sensitizing agent