Table 3.

Details of accepted trials

Article	DS	Patient character-istics/potential confounders*	GIC treatment group					Amalgam treatment group					Outcome measure	Evaluation		Dentition/Teeth/Restoration	Study period
			Type of material	BSL	N	n	LTF	Type of material	BSL	N	n	LTF		Criteria	Method
Welbury et al., 1991 [20]	01	[1]	Ketac Fil	119	51	7	68	Amalcap	119	51	11	68	Recurrent caries	USPHS	Clinical examination	Primary/Molars/Class I & II	5 years
Östlund et al., 1992 [21]	02	[2]	Chem Fil	25	10	0	15	ANA 2000	25	23	1	2	Recurrent caries	USPHS	Clinical examination	Primary/Molars/Class II	3 years
Taifour et al., 2002 [22]	03	[3]	Fuji IX/ Ketac Molar	610	475	9	135	Avalloy	425	331	11	94	Caries on margin	ART	Clinical examination	Primary/Molars/Single surface	3 years
	04			478	106	4	372		380	84	9	296				Primary/Molars/Multiple surface
Mandari et al., 2003 [23]	05	[4]	Fuji II	223	173	3	50	ANA 2000	207	162	16	45	Recurrent caries	Modified USPHS	Clinical examination	Permanent/Molars/Single surface	6 years
Frencken et al., 2007 [24]	06	[5]	Fuji IX/ Ketac Molar	487	153	11	334	Avalloy	403	108	15	295	Caries on margin	ART	Clinical examination	Permanent/Molars/Single surface	6.3 years
Daou et al., 2009 [25]	07	[6]	Fuji IX	35	33	4	2	Permite C	38	36	1	2	Recurrent caries	USPHS	Clinical examination	Primary/Molars/Class I & II	1 year
	08			35	23	3	12		38	21	3	17					2 years
Mandari et al., 2001 [27]	09	[7]	Fuji II	223	211	9	12	ANA 2000	207	196	15	11	Recurrent caries	Modified USPHS	Clinical examination	Permanent/Molars/Single surface	2 years
Yu et al., 2004 [28]	10	[8]	Fuji IX/ Ketac Molar-aplicap	45	37	0	8	GK amalgam	32	23	0	9	Recurrent caries	ART	Clinical examination	Primary/Molars/Single surface	1 year
	11			45	29	0	16		32	18	0	14					2 years
Svanberg, 1992 [29]	12	[9]	Ketac Silver	18	14	0	4	Disper-salloy	18	14	3	4	Recurrent caries	SNBHW	Clinical examination	Permanent/Molars & Premolars/GIC = Tunnel/Amalgam = Class II	3 years
	13			11	11	3	0		11	11	9	0	Caries progression		Probing & Bitewing	Permanent/approximal adjacent surfaces
Qvist et al., 1997 [26]	14	[10]	Ketac Fil	515	334	11	181	Disper-salloy	543	306	17	237	Recurrent caries	DPDHS	Clinical examination	Primary/Class I, II and III/V	3 years
	15			127	105	25	22		127	94	47	33	Caries progression			Primary & permanent/approximal adjacent surfaces (sound or arrested caries)
	16			156	120	25	36		183	129	47	54	Caries progression			Primary/approximal adjacent surfaces (carious or active lesion)
	17			156	120	66	36		183	129	78	54	No caries regression			Primary/approximal adjacent surfaces (carious or active lesion)

DS = Dataset number; BSL = Number of teeth at baseline; N = Number of teeth evaluated; n = Number of teeth with caries, LTF = Loss-to-follow-up; USPHS = United States Public Health Service criteria; ART = Criteria for atraumatic restorative treatment; SNBHW = Criteria according to the Swedish National Board for Health and Welfare; DPHS = Danish Public Dental Health Service criteria.

* Potential confounders = Reported fluoride exposure; high-sugary diet; poor oral hygiene; high past caries experience.

Patient characteristics:

[1] Split-mouth trial. 76 patients, age 5 - 11 years; patients attending the Department of Child Dental Health at Newcastle Dental Hospital (UK) for routine restorative care; subjects were admitted to the trials if they required at least 1 pair of restorations in their deciduous molar dentition; paired cavities either Class I or II, if possible in the same tooth type; restoration always in different quadrants per pair; any cavity was suitable for inclusion; a cavity was excluded if it could only be satisfactory restored using a stainless steel crown; restorations placed between October 1982 and March 1987;caries removal by drill.

Potential confounders reported: none.

[2] Partial split-mouth trial. 56 patients, age 4-6 years regularly treated at one Public Dental Service clinic in Jönköping, Sweden who showed manifest caries lesion on the mesial surface of a 2^nd primary molar; lesion not atypical or extended into buccal or lingual tooth surfaces; lesion completely surrounded by healthy enamel and should not reach the pulp; caries removal by drill;

Potential confounders reported: none.

[3] Parallel group trial. 835 patients, age 6-7 years from Damascus, Syria; with dentinal lesions with an opening wide enough for the smallest excavator to enter (diameter = 0.9 mm, without pulp involvement; size of restorations varied from small to large; dental caries prevalence 85%; mean dmfts and dmft scores of molars plus canines 9.0 and 4.4, respectively; GIC restorations placed after caries removal by hand excavation (ART).

Potential confounders reported: High past caries experience.

[4] Split-mouth trial. 152 patients from a cohort of grade 3-5 pupils, mean age 11 years in need of 2 or more restorations; from urban and rural schools near Dar es Salaam, Tanzania; selection criteria concerned dentine lesions in the occlusal surface that showed no evidence of pulpal involvement; pupils needed to have a dentine lesion present in contralateral permanent molars; infected dentine was removed with slow-speed drill and excavators or by hand excavation with use of Caridex.

Potential confounders reported: none.

[5] Parallel group trial. A total of 108 children of the ART group (GIC) and 84 children of the amalgam group were examined at evaluation year 6.3 - from Damascus, Syria; mean age 13.8 years; high risk for dentine lesion development (mean DMFT score 5.5); the mean DMFT and DMFS scores of the children in the ART group were 5.5 (SD 3.0) and 8.2 (SD 5.4) respectively; the mean DMFT and DMFS scores of the children in the amalgam group were 6.0 (SD 1/4 3.3) and 9.4 (SD 6.4); there was no statistically significant difference in caries scores between the children of the two groups (P > 0.05); the mean plaque score for the children in the ART and amalgam group were 1.3 (SD 0.58) and 1.2 (SD 0.52), respectively (see also [3]).

Potential confounders reported: Poor oral hygiene; high past caries experience.

[6] Partial split-mouth trial. 45 girls 6-8 years old from a private school (boarding and regular school) in Beirut, Lebanon; from a low socio-economic background with their first and second primary molars requiring new Class I or Class II restorations; specific criteria included vital teeth with normal appearance and morphology, and teeth with or without adjacent teeth; the children routinely (before and during study) received information and instructions to improve their oral hygiene, and had two dental examinations per year; criteria for exclusion from the study: patients having behavioural problems, patients with general health problems, patients with poor oral hygiene, molars requiring pulpotomy or pulpectomy; caries removal with drill;

Potential confounders reported: none.

[7] Split-mouth trial. see [4]

[8] Split-mouth trial. 60 Chinese children with mean age 7.4 (SD 1.24) years; 27 boys, 33 girls in Bejing; caries removal for GIC (ART) restorations by hand excavation or drill.

Potential confounders reported: none.

[9] Split-mouth trial. 18 caries-active patients, aged 13-16 years; from the regular clientele visiting one of the dental clinics of the Public Dental Health Service in Kronoberg, Sweden; with proximal primary, early carious lesions on contralateral posterior teeth needed restorative treatment; lesion extending into dentin; have progressed into deeper zone since preceding information; caries removal by drill.

Potential confounders reported: none.

[10] Partial split-mouth trial. 666 children, from 3 to 13 years of age within the Danish Public Dental Health Service in the municipalities of Vaerløse and Hillerød, Denmark. The caries experience among children and adolescents in the two municipalities is below the national average; caries removal by drill.

Potential confounders reported: Low past caries experience