Intervention(s) and comparator(s)	Sex [female %]	Age [mean years SD)/range]	BMI [mean kg/m² (SD)]	Weight [mean kg (SD)]	Co‐medications / Co‐interventions [%]	Co‐morbidities [N, % or as stated]
Aasheim 2009	Laparoscopic Roux‐en‐Y gastric bypass	74	35 (7)	54.8 (3.2)	162 (24)	‐	N (rounded %): type 2 diabetes 6 (19), joint pain 16 (52), depression 5 (16), hypertension 8 (26), asthma 8 (26), urinary incontinence 5 (16), sleep apnoea 5 (16), GERD 5 (16), diabetes mellitus 5 (16), hypothyroidism 3 (10), gallstones 2 (7), hyperlipidaemia 0 (0), gout 1 (3)
	Laparoscopic biliopancreatic diversion with duodenal switch	66	36 (5)	55.2 (3.5)	162 (20)	‐	N (rounded %): type 2 diabetes 6 (21), joint pain 13 (45), depression 12 (41), hypertension 8 (28), asthma 5 (17), urinary incontinence 7 (24), sleep apnoea 6 (21), GERD 4 (14), diabetes mellitus 3 (10), hypothyroidism 3 (10), gallstones 1 (3), hyperlipidaemia 3 (10), gout 1 (3)
	all:	70	‐	‐	‐	‐
Angrisani 2007	Laparoscopic Roux‐en‐Y gastric bypass	83	34.1 (8.9)	43.8 (4.1)	118.2 (13.2)	‐	N: 2 hyperlipaemia, 1 hypertension, 1 type 2 diabetes
	Laparoscopic adjustable gastric banding	81	33.8 (9.1)	43.4 (4.2)	117.1 (12.8)	‐	N: 3 hypertension, 1 sleep apnoea
Demerdash 2013	Laparoscopic Roux‐en‐Y gastric bypass	89a	39 (4.5)	46.2 (2.56)	142.7 (22.6)	‐	‐
	Laparoscopic adjustable gastric band	83	37 (6)	45.8 (2.7)	138.7 (22.98)	‐	‐
Dixon 2008	Laparoscopic gastric banding in addition to the conventional therapy	50	46.6 (7.4)	37.0 (2.7)	105.6 (13.8)	n/N (%)b: Metformin: 28/29 (97); Other hypoglycaemic agents: 9/29 (31); Insulin: 1/29 (3); Antihypertensive agents: 20/29 (69); Lipid‐lowering agents: 12/29 (41)	N (%): type 2 diabetes 30 (100), hypertension 28 (93), metabolic syndrome 29 (97), coronary artery disease 0 (0)
	Conventional therapy	57	47.1 (8.7)	37.2 (2.5)	105.9 (14.2)	n/N (%)b: Metformin: 26/26 (100); Other hypoglycaemic agents: 8/26 (31); Insulin: 0/26 (0); Antihypertensive agents: 15/26 (58); Lipid‐lowering agents: 8/26 (31)	N (%): type 2 diabetes 30 (100), hypertension 27 (90), metabolic syndrome 29 (97), coronary artery disease 1 (3)
Dixon 2012	Laparoscopic adjustable gastric banding	43	47.4 (8.8)	46.3 (6.0)	134.9 (22.1)	‐	N (%): obstructive sleep apnoea 30 (100), hypertension 15 (50), diabetes 10 (33), depression 12 (40), metabolic syndrome 19 (63)
	Conventional therapy	40	50.0 (8.2)	43.8 (4.9)	126.0 (19.3)	‐	N (%): obstructive sleep apnoea 30 (100), hypertension 17 (57), diabetes 10 (33), depression 11 (37), metabolic syndrome 24 (80)
Hedberg 2012	Biliopancreatic diversion with duodenal switch	50	40.2 (9.5)	54.5 (6.7)	‐	N (%): oral diabetes medication 6 (25), insulin 1 (4), any diabetes medication 7 (29)	N (%): hypertension 6 (25), hyperlipidaemia 0 (0), sleep apnoea 4 (17)
	Roux‐en‐Y gastric bypass	43	37.9 (10.4)	54.5 (5.6)	‐	N (%): oral diabetes medication 1 (4), insulin 0 (0), any diabetes medication 1 (4)	N (%): hypertension 7 (30), hyperlipidaemia 0 (0), sleep apnoea 3 (13)
	all:	47	39.1 (9.9)	54.5 (6.1)	‐
Himpens 2006	Laparoscopic gastric banding	83	median 36 (20‐61)	median 37 (30‐47)	‐	‐	N (%): GERD requiring proton pump inhibitor 6 (15)
	Laparascopic isolated sleeve gastrectomy	78	median 40 (22‐65)	median 39 (30‐53)	‐	‐	N (%): GERD requiring proton pump inhibitor 8 (20)
Ikramuddin 2013	Laparoscopic Roux‐en‐Y gastric bypass + lifestyle programme with medical management	63	49 (9)	34.9 (3.0)	98.8 (14.0)	insulin 62, other glycaemic medicines 87, dyslipidaemia medicines 65, blood pressure medicines 68	% inferred from inclusion criteria: type 2 diabetes 100, elevated HbA1c (> 8%) 100
	Lifestyle programme with medical management	57	49 (8)	34.3 (3.1)	97.9 (17.0)	insulin 43, other glycaemic medicines 95, dyslipidaemia medicines 68, blood pressure medicines 73	% inferred from inclusion criteria: type 2 diabetes 100, elevated HbA1c (> 8%) 100
	all:			34.6 (3.1)
Karamanakos 2008 (including Kehagias 2011)	Laparoscopic Roux‐en‐Y gastric bypass	73	36 (8.4)	45.8 (3.7)	123.1 (13.9)	‐	N (%): hypertension 5 (17), type 2 diabetes 5 (17), impaired glucose tolerance 5 (17), HDL < threshold 4 (13), LDL > threshold 10 (33), triglycerides > threshold 5 (17), obstructive sleep apnoea 3 (10), GERD 5 (17), degenerative arthritis 6 (20), menstrual irregularities 7 (23), ≥ 1 obesity‐related co‐morbidity 23 (77)
	Laparoscopic sleeve gastrectomy	73	33.7 (9.9)	44.9 (3.4)	126.9 (18.0)	‐	N (%): hypertension 4 (13), type 2 diabetes 5 (17), impaired glucose tolerance 5 (17), HDL < threshold 3 (10), LDL > threshold 8 (27), triglycerides > threshold 3 (10), obstructive sleep apnoea 6 (20), GERD 2 (7), degenerative arthritis 5 (17), menstrual irregularities 7 (23), ≥ 1 obesity‐related co‐morbidity 20 (67)
	all:	73	‐	‐	‐
Keidar 2013	Laparoscopic Roux‐en‐Y gastric bypass	42c	51.45 (8.3)	42 (4.8)	118.04 (16.5)	oral hypoglycaemics: 63%; insulin: 21%	Type 2 diabetes: 100%
	Laparoscopic sleeve gastrectomy	50c	47.7 (11.7)	42.5 (5.2)	117.9 (17.8)	oral hypoglycaemics: 50%; insulin: 22%	Type 2 diabetes: 100%
	all:	46
Lee 2011	Simplified laparoscopic mini‐gastric bypass with duodenum exclusion	‐	‐	‐	‐	‐	%: poorly‐controlled type 2 diabetes 100
	Laparoscopic sleeve gastrectomy without duodenum exclusion	‐	‐	‐	‐	‐	%: poorly‐controlled type 2 diabetes 100
Liang 2013	Usual care	33.3	51.75 (6.70)	30.34 (1.96)d	81.31 (4.97)	%: insulin therapy in combination with (unspecified) oral agents 100	%: type 2 diabetes 100, hypertension 100
	Usual care + exenatide	29.4	50.94 (5.89)	30.28 (1.44)	81.76 (3.67)	%: insulin therapy in combination with (unspecified) oral agents 100	%: type 2 diabetes 100, hypertension 100
	Laparoscopic Roux‐en‐Y gastric bypass	29	50.81 (5.44)	30.48 (0.94)	81.97 (3.53)	%: insulin therapy in combination with (unspecified) oral agents 100	%: type 2 diabetes 100, hypertension 100
	all:	‐	‐	30.3 [25.0‐34.0]	‐	‐
Mingrone 2012	Gastric bypass	60	43.90 (7.57)	44.85 (5.16)	129.84 (22.58)	‐	%: type 2 diabetes 100, elevated HbA1c (≥ 7%) 100
	Medical therapy	50	43.45 (7.27)	45.62 (6.24)	136.40 (21.94)	‐	%: type 2 diabetes 100, elevated HbA1c (≥ 7%) 100
	all:	53	‐	‐	‐
Nguyen 2009	Laparoscopic Roux‐en‐Y gastric bypass	77.4	41.4 (11.0)	47.5 (5.5)	133 (21)	‐	%: diabetes 20.7, hypertension 38.7, previous abdominal surgery 45.9
	Laparoscopic adjustable gastric banding	75.6	45.8 (9.8)	45.5 (5.4)	129 (21)	‐	%: diabetes 26.7, hypertension 51.1, previous abdominal surgery 47.7
Nogues 2010	Laparascopic Roux‐en‐Y gastric bypass	100	45.86 (8.6)e	43.1 (3.9)e	116.7 (5.5)	N (%): metformin 2 (28.6)	N (rounded %): hypertension 5 (71), diabetes mellitus 2 (29), dyslipidaemia 5 (71), arthropathy 4 (57), GERD 0 (0), urinary incontinence 3 (43), depression 4 (57), obstructive sleep apnoea 2 (29), insulin resistance 6 (86)
	Laparoscopic sleeve gastrectomy	100	49.63 (9.6)	43.5 (3.2)	108.9 (6.3)	N (%): metformin 2 (25.0)	N (rounded %): hypertension 7 (88), diabetes mellitus 2 (25), dyslipidaemia 5 (63), arthropathy 6 (75), GERD 1 (13), urinary incontinence 6 (75), depression 7 (88), obstructive sleep apnoea 2 (25), insulin resistance 4 (50)
	all:	100	47.8 (9.0)	43.3 (3.4)	‐
O'Brien 2006	Laparoscopic adjustable gastric band	75	41.8 (6.4)	33.7 (1.8)	96.1 (11.2)	‐	%: hypertension 22.5, metabolic syndrome 37.5, coronary artery disease 0
	Intensive non‐surgical programme	77.5	40.7 (7.0)	33.5 (1.4)	93.6 (11.9)	‐	%: hypertension 17.5, metabolic syndrome 37.5, coronary artery disease 0
Paluszkiewicz 2012	Roux‐en‐Y gastric bypass	64	43.9 (10.8)	48.6 (5.4)	137.7 (17.7)	‐	N (rounded %): hypertension 30 (83), type 2 diabetes 14 (39), dyslipidaemia 31 (86)
	Laparoscopic sleeve gastrectomy	72	44.9 (10.6)	46.1 (5.9)	130.7 (15.5)	‐	N (rounded %): hypertension 25 (69), type 2 diabetes 10 (28), dyslipidaemia 31 (86)
	all:	68	‐	‐	‐
Peterli 2012	Laparoscopic Roux‐en‐Y gastric bypass	72	42.1 (11.2)	44.2 (5.3)	124.8 (19.8)	‐	%: hypertension 59, type 2 diabetes 26, dyslipidaemia 51, obstructive sleep apnoea 42, GERD 46, back/joint arthralgia 68, depression 11
	Laparoscopic sleeve gastrectomy	72	43.0 (11.1)	43.6 (5.3)	123.5 (19.4)	‐	%: hypertension 63, type 2 diabetes 24, dyslipidaemia 67, obstructive sleep apnoea 48, GERD 44, back/joint arthralgia 61, depression 20
	all:	72	43.0 (5.3)	44 (11.1)	‐
Praveen Raj 2012	Laparoscopic duodenojejunal bypass with sleeve gastrectomy	64	39.5	48.28 (3.80)	‐	‐	N (%): type 2 diabetes 20 (71), hypertension 10 (36)
	Laparoscopic Roux‐en‐Y gastric bypass	55	43.5	49.29 (3.63)	‐	‐	N (%): type 2 diabetes 16 (55), hypertension 12 (41)
	all:	60	‐	‐	‐
Schauer 2012	Laparoscopic Roux‐en‐Y gastric bypass	58	48.3 (8.4)	37.0 (3.3)	106.7 (14.8)	N (%): insulin 22 (44)	N (%): type 2 diabetes 50 (100), elevated HbA1c (> 7%) 50 (100), metabolic syndrome 45 (90), dyslipidaemia history 44 (88), hypertension history 35 (70)
	Laparoscopic sleeve gastrectomy	78	47.9 (8.0)	36.2 (3.9)	100.8 (16.4)	N (%): insulin 22 (44)	N (%): type 2 diabetes 50 (100), elevated HbA1c (> 7%) 50 (100), metabolic syndrome 47 (94), dyslipidaemia history 40 (80), hypertension history 30 (60)
	Intensive medical therapy	62	49.7 (7.4)	36.8 (3.0)	106.5 (14.7)	N (%): insulin 22 (44)	N (%): type 2 diabetes 50 (100), elevated HbA1c (> 7%) 50 (100), metabolic syndrome 46 (92), dyslipidaemia history 36 (84), hypertension history 26 (60)
Sharma 2013	Laparoscopic gastric imbrication	‐	40.5	44.7 (6.1)	‐	‐	‐
	Laparoscopic sleeve gastrectomy	‐	39.9	44.0 (7.8)	‐	‐	‐
Vix 2013	Laparoscopic Roux‐en‐Y gastric bypass	87	35.23 (9.37)	47.09 (5.64)	129.58 (21.17)	‐	%: type 2 diabetes 8.9, hypertension 37.8, sleep apnoea syndrome 20.0, vitamin D deficiency 85.7, hypercholesterolaemia 26.7, abnormal LDL 11.1, abnormal HDL 6.7, abnormal triglycerides 17.8, hyperparathyroidism 24.0
	Laparoscopic sleeve gastrectomy	78	35.13 (9.7)	45.57 (4.79)	128.68 (18.27)	‐	%: type 2 diabetes 7.3, hypertension 21.8, sleep apnoea syndrome 9.1, vitamin D deficiency 84.6, hypercholesterolaemia 27.3, abnormal LDL 9.1, abnormal HDL 5.5, abnormal triglycerides 27.3, hyperparathyroidism 28.13
"‐" denotes not reported aUnclear whether data are based on 16 or 18 participants in this group ‐ the data extracted here assume 18 (16 females and 2 males) bN = number completing study cBaseline characteristics of per protocol population only presented dThis was reported as 30.34 in table 1 and 30.94 in table 2 eData here are from the Nogues paper ‐ those from the associated Ramon paper are slightly different GERD: gastroesophageal reflux disease; HbA1c: glycosylated haemoglobin A1c: HDL: high density lipoprotein; LDL: low density lipoprotein