Table 1.

A summary of included studies on non-surgical approaches in treating dentin caries in preschool children

Author/year	Study type/duration	Subjects	Intervention	Outcome assessment	Results/conclusion	Comment/quality assessment
Zhi et al., 2012 [15]	RCT	212 children	Gp 1: 38% SDF once a year	Clinical examination criteria:	At 1 year, the caries arrest rates of Gps 1, 2 and 3 were 37%, 53% and 29%, respectively.	Well-planned study design (random allocation, adequate sample size) Dropout rate was not high (15%). Confounding factors were taken into consideration.
	2 years	aged 3–4 years with active caries in primary teeth	Gp 2: 38% SDF twice a year	Active: lesion easily penetrated by probe
			Gp 3: Flowable GI filling once a year	Arrested: smooth, hard surfaces when probing	At 2 years, the caries arrest rates of three groups were 79%, 91% and 82%, respectively.
				But for Gp 3 also lesions that were totally covered with GI	Effect of annual SDF and GI application on arresting caries did not differ significantly.	However, blinding of outcome assessment (between Gp 3 and Gp 1, 2) is impossible.
					SDF application twice a year increased caries arrest rates.	Low risk of bias
						Study quality (ADA): good
dos Santos et al., 2012 [14]	RCT	91 children aged 5–6 years with caries in primary teeth	Gp 1: interim restorative treatment with GI filling without caries removal	Clinical examination criteria:	At 12 months, the success rate of SDF was higher than interim restoration with GI filling (67% vs. 39%).	No details about the random allocation and attrition rate. Non-blinded study and duration of study was short.
	1 year
				Active: lesion easily penetrated by probe
			Gp 2: 30% SDF	Arrested: smooth, hard surfaces when probing	SDF was more effective than interim restoration with GI for arresting caries in primary teeth.	High risk of bias
						Study quality (ADA): poor
Chu et al., 2002 [13]	RCT	375 children	Gp 1: excavation plus 38% SDF once a year	Clinical examination criteria:	SDF groups (Gp 1, 2) had higher caries arrest rates than those of NaF groups (Gp 3, 4) and control. The respective mean numbers of arrested caries tooth surfaces in the five groups were 2.5, 2.8, 1.5, 1.5 and 1.3, respectively.	Ethical concern regarding the negative control group (no treatment)
	30 months	(aged 3–5 years)
		Upper anterior primary teeth	Gp 2: 38% SDF once a year	Arrested caries: cavity with hard floor and walls
						Only anterior primary teeth were involved. The generalizability of the results to posterior teeth was limited.
			Gp 3: excavation plus 5% NaF 4 times a year
			Gp 4: 5% NaF 4 times a year
			Gp 5: Control (no treatment)		SDF was effective in arresting dentin caries.	Low risk of bias
						Study quality (ADA): good
Lo et al., 1998 [9]	Longitudinal study	289 children aged 3–6 years	Gp 1: Regular oral health education session and daily tooth brushing with 1000 ppm fluoridated toothpaste	Clinical examination criteria:	Significant difference between the mean no. of arrested caries in Gp 1 and 2 which was 1.8 and 1.1, respectively.	Low attrition rate over 3 years. Potential confounders were evaluated. Although no random allocation was performed, comparable groups were assembled initially and maintained throughout the study.
	3 years	(168 children in intervention group, 121 children in the control group)		Arrested caries: dark brown to black in color with hard surface
					At 3 years, 28% of the active dentin caries in Gp 1 children had become arrested while 12% of the caries were arrested in the control.
			Gp 2: Control (no intervention)
						Moderate risk of bias
						Study quality (ADA): good