**3. Etiology of secondary caries**

#### **3.1 Microbiology of secondary caries**

Dental caries is determined by the dynamic balance between pathological factors that lead to demineralization and protective factors that lead to remineralization [Featherstone, 2004]. As a major pathological factor, oral bacteria, especially acidogenic bacteria, can dissolve the tooth mineral. Those acidogenic bacteria are also aciduric and can live preferentially under acid conditions [Loesche, 1986].

Hitherto, it is unclear about the microbiology of secondary caries yet. Although secondary caries is described alike primary caries in histopathology, whether the etiology of secondary caries is the same as that of primary caries is a matter in dispute. Kidd et al. found no

Secondary Caries 409

grow on blood agar [Thomas et al., 2008]. In the past decade, the detection of *A. Odontolyticus* and *Candida spp*. has caused the serious concern to researchers. It has been found that *Candida albicans* can dissolve hydroxyapatite in a liquid culture at a 20-fold higher rate than *S. mutants*, despite the much lower growth rate [Nikawa et al., 2003]. Klinke et al. assumed that *Candida albicans* might make a significant contribution to caries pathogenesis in caries-active children, and it could be taken into account *Candida albicans* as an appraisal of caries pathogenicity [Klinke et al, 2009]. Besides, it should be noted that some people may have serious caries activities without S. mutants dominating in dental plaque. Therefore, further research need to be carried out to determine the microbiology of secondary caries, such as the role of *S. mutants*, *A. Odontolyticus* and *Candida spp.* in the development of secondary caries and the relationship between restorations and

Microleakage refers to the clinically undetectable leakage between the cavity wall and the filling [Kidd, 1976]. Irie et al. found that a gap of 6-10µm formed immediately even after applying an acid etch and a bonding agent [Irie et al., 2002]. Iwami et al. have confirmed that any restorative material can completely eliminate the microleakage between restoration and the cavity wall [Iwami et al., 2005], supported by other researches [Irie and Suzuki, 1999; Huang et al., 2002; Piwowarczyk et al, 2005]. A study in vitro showed that there was no significant difference in the degree of microleakage between conventional caries removal and chemo-mechanical removal [Mousavinenasab and Jafary, 2004]. Those above all show

The microspace between the restoration wall and tooth can allow salivary pellicle accumulation and bacterial invasion [González-Cabezas et al., 1999; González-Cabezas et al., 2002; Splieth et al., 2003]. In a sense, it provides a favorable environment for the oral bacteria, especially cariogenic bacteria, such as S. mutants and Lacotobacilli, to demineralize the tooth structure along the cavity wall, as long as conditions are adequate and suitable. The histopathological appearance of the wall lesion in the secondary caries is also explained by hydrogen ions due to the diffusion of bacteria into the space between restoration and cavity wall and its acdiogenic activities afterwards [Hals and Nernaes, 1971]. So microleakage has been considered as a potential predictor for secondary caries and has caused serious concern to many researchers. In some article, the wall lesion was described as

Up to now, there has been no conclusive statement about the relationship between microleakage and secondary caries. Several studies in vitro have shown a positive relationship between the two things. Jørgensen and Wakumoto in a 1968 research found that there was an increasing likelihood of secondary caries with the increasing size of the microspace [Jørgensen and Wakumoto, 1968], which is in agreement with that reported by some other researches [Goldbeg et al., 1981; Dérand et al., 1991]. In a recent in vitro study on relationship of gap size and secondary caries, the findings suggested that the gap size between tooth and restoration affected the development of secondary caries along the cavity wall [Totiam et al., 2007], for which the rationale was that bigger gaps would provide necessary space for bacterial colonization and enough nutrients for cariogenic microorganisms leading to the creation of larger wall lesions. On the other hand, within

microorganism of secondary caries.

us that microleakage is inevitable.

**3.2 The relevance of microleakage to secondary caries** 

the consequence of microleakge [Diercke et al., 2009].

significant differences between the microflora in samples from cavity walls involving primary and secondary caries next to the amalgam [Kidd et al., 1993]. However, Thomas et al. investigated bacterial composition in relation to primary and secondary caries via an in situ model, and found a phenomenon of higher proportion of caries-associated bacteria on composite surfaces. Then they indicated that the microbiology on the surface of the primary caries differs from that on the surface of lesion around composite, and secondary caries around composite may differ from the primary lesions process [Thomas et al., 2008]. In addition, some studies focused on the ecology under the restorations. Mejàre et al. found the bacterial colonization beneath composite similar to that observed in dental plaque mainly including Streptococci and Actinomyces spp. [Mejàre et al., 1979]. Nevertheless, according to the experiment conducted by Splieth et al., it was the other way around [Splieth et al., 2003]. They compared the microbial spectrum under composite and amalgam fillings with special attention to the anaerobic flora. The results showed that bacterial composition under amalgam was similar to the flora of carious dentin and carious plaque, with anaerobic and facultative anaerobic gram-positive rods dominating. On the contrary, huge amounts of Bacteroides and Prevotella spp. were detected under many composite fillings, similar to the microflora of infected root canals with potentially pulpopathogenic microbes. Thus, the study suggested that the types of restorative materials seemed to have an effect on the composition of the microflora on the surface of secondary caries, and that beneath the restorations, and then the differences might exist between the microbial flora of secondary caries and primary caries. In this study, it was indicated that inadequate composite fillings might stimulate the growth of cariogenic as well as obligate anaerobic and potentially pulpopathogenic bacteria. This could be explained as follows: 1) The microspace between the restoration and the cavity floor favors the obligate anaerobic, and then leads to the detection of those bacteria; 2) It is not surprising to discover many obligate anaerobic normally colonize in human oral, even in oral of people without obvious endodontic diseases. 3) It doesn't mean that people without clinical symptoms of toothaches or pulpitis don't have chronic or arrested tooth diseases, so it is possible to detect those anaerobic bacteria. However, existence does not mean participation, so it is necessary to certify the participation of those obligate anaerobic in the progress of secondary caries in further studies.

According to the viewpoint of Marsh that any species with the ability producing acids and tolerating the cariogenic environment can contribute to the dental caries process [Marsh, 2006]. *Streptococci mutans (S. mutans), Lactobacilli* and *Actinomyces naeslundii* have been used by various models in vitro studying secondary caries for a long time. *S. mutants* and lactobacilli can produce a series of acid and stay in a low pH environment for a long time, leading to demineralization of teeth and caries lesion. It has been shown that the three bacteria were widely present and might play an important role in the development of secondary caries around amalgam [González-Cabezas, 1999]. However, in a recent in situ study, *S. mutants* were not detected in each sample, but *Lactobacilli*. Meanwhile, *A. Odontolyticus* and *Candida spp*. were also found in most samples [Thomas et al., 2008]. In addition, in recent years, Beighton put forward a point of view — *S. mutants* might be good makers of secondary caries but not necessarily the etiological agents [Beighton, 2005]. The experiment by Thomas et al. described before, in which *S. mutants* was not found in every sample, but *Lactobacilli* and *A. Odontolyticus*, seemed to support Beighton's view. And scientists conjectured that there might be unknown caries-associated bacteria, which can not

significant differences between the microflora in samples from cavity walls involving primary and secondary caries next to the amalgam [Kidd et al., 1993]. However, Thomas et al. investigated bacterial composition in relation to primary and secondary caries via an in situ model, and found a phenomenon of higher proportion of caries-associated bacteria on composite surfaces. Then they indicated that the microbiology on the surface of the primary caries differs from that on the surface of lesion around composite, and secondary caries around composite may differ from the primary lesions process [Thomas et al., 2008]. In addition, some studies focused on the ecology under the restorations. Mejàre et al. found the bacterial colonization beneath composite similar to that observed in dental plaque mainly including Streptococci and Actinomyces spp. [Mejàre et al., 1979]. Nevertheless, according to the experiment conducted by Splieth et al., it was the other way around [Splieth et al., 2003]. They compared the microbial spectrum under composite and amalgam fillings with special attention to the anaerobic flora. The results showed that bacterial composition under amalgam was similar to the flora of carious dentin and carious plaque, with anaerobic and facultative anaerobic gram-positive rods dominating. On the contrary, huge amounts of Bacteroides and Prevotella spp. were detected under many composite fillings, similar to the microflora of infected root canals with potentially pulpopathogenic microbes. Thus, the study suggested that the types of restorative materials seemed to have an effect on the composition of the microflora on the surface of secondary caries, and that beneath the restorations, and then the differences might exist between the microbial flora of secondary caries and primary caries. In this study, it was indicated that inadequate composite fillings might stimulate the growth of cariogenic as well as obligate anaerobic and potentially pulpopathogenic bacteria. This could be explained as follows: 1) The microspace between the restoration and the cavity floor favors the obligate anaerobic, and then leads to the detection of those bacteria; 2) It is not surprising to discover many obligate anaerobic normally colonize in human oral, even in oral of people without obvious endodontic diseases. 3) It doesn't mean that people without clinical symptoms of toothaches or pulpitis don't have chronic or arrested tooth diseases, so it is possible to detect those anaerobic bacteria. However, existence does not mean participation, so it is necessary to certify the participation of those obligate anaerobic in the progress of

According to the viewpoint of Marsh that any species with the ability producing acids and tolerating the cariogenic environment can contribute to the dental caries process [Marsh, 2006]. *Streptococci mutans (S. mutans), Lactobacilli* and *Actinomyces naeslundii* have been used by various models in vitro studying secondary caries for a long time. *S. mutants* and lactobacilli can produce a series of acid and stay in a low pH environment for a long time, leading to demineralization of teeth and caries lesion. It has been shown that the three bacteria were widely present and might play an important role in the development of secondary caries around amalgam [González-Cabezas, 1999]. However, in a recent in situ study, *S. mutants* were not detected in each sample, but *Lactobacilli*. Meanwhile, *A. Odontolyticus* and *Candida spp*. were also found in most samples [Thomas et al., 2008]. In addition, in recent years, Beighton put forward a point of view — *S. mutants* might be good makers of secondary caries but not necessarily the etiological agents [Beighton, 2005]. The experiment by Thomas et al. described before, in which *S. mutants* was not found in every sample, but *Lactobacilli* and *A. Odontolyticus*, seemed to support Beighton's view. And scientists conjectured that there might be unknown caries-associated bacteria, which can not

secondary caries in further studies.

grow on blood agar [Thomas et al., 2008]. In the past decade, the detection of *A. Odontolyticus* and *Candida spp*. has caused the serious concern to researchers. It has been found that *Candida albicans* can dissolve hydroxyapatite in a liquid culture at a 20-fold higher rate than *S. mutants*, despite the much lower growth rate [Nikawa et al., 2003]. Klinke et al. assumed that *Candida albicans* might make a significant contribution to caries pathogenesis in caries-active children, and it could be taken into account *Candida albicans* as an appraisal of caries pathogenicity [Klinke et al, 2009]. Besides, it should be noted that some people may have serious caries activities without S. mutants dominating in dental plaque. Therefore, further research need to be carried out to determine the microbiology of secondary caries, such as the role of *S. mutants*, *A. Odontolyticus* and *Candida spp.* in the development of secondary caries and the relationship between restorations and microorganism of secondary caries.

#### **3.2 The relevance of microleakage to secondary caries**

Microleakage refers to the clinically undetectable leakage between the cavity wall and the filling [Kidd, 1976]. Irie et al. found that a gap of 6-10µm formed immediately even after applying an acid etch and a bonding agent [Irie et al., 2002]. Iwami et al. have confirmed that any restorative material can completely eliminate the microleakage between restoration and the cavity wall [Iwami et al., 2005], supported by other researches [Irie and Suzuki, 1999; Huang et al., 2002; Piwowarczyk et al, 2005]. A study in vitro showed that there was no significant difference in the degree of microleakage between conventional caries removal and chemo-mechanical removal [Mousavinenasab and Jafary, 2004]. Those above all show us that microleakage is inevitable.

The microspace between the restoration wall and tooth can allow salivary pellicle accumulation and bacterial invasion [González-Cabezas et al., 1999; González-Cabezas et al., 2002; Splieth et al., 2003]. In a sense, it provides a favorable environment for the oral bacteria, especially cariogenic bacteria, such as S. mutants and Lacotobacilli, to demineralize the tooth structure along the cavity wall, as long as conditions are adequate and suitable. The histopathological appearance of the wall lesion in the secondary caries is also explained by hydrogen ions due to the diffusion of bacteria into the space between restoration and cavity wall and its acdiogenic activities afterwards [Hals and Nernaes, 1971]. So microleakage has been considered as a potential predictor for secondary caries and has caused serious concern to many researchers. In some article, the wall lesion was described as the consequence of microleakge [Diercke et al., 2009].

Up to now, there has been no conclusive statement about the relationship between microleakage and secondary caries. Several studies in vitro have shown a positive relationship between the two things. Jørgensen and Wakumoto in a 1968 research found that there was an increasing likelihood of secondary caries with the increasing size of the microspace [Jørgensen and Wakumoto, 1968], which is in agreement with that reported by some other researches [Goldbeg et al., 1981; Dérand et al., 1991]. In a recent in vitro study on relationship of gap size and secondary caries, the findings suggested that the gap size between tooth and restoration affected the development of secondary caries along the cavity wall [Totiam et al., 2007], for which the rationale was that bigger gaps would provide necessary space for bacterial colonization and enough nutrients for cariogenic microorganisms leading to the creation of larger wall lesions. On the other hand, within

Secondary Caries 411

As secondary caries is one of the major reasons for restoration replacement, a large number of clinical dentists and scientists have placed great emphasis on preventing or slowing down the procession of secondary caries lesion from many aspects, so as to increase clinical restoration durability. Secondary caries, the same as other types of dental caries, is determined by the dynamic balance between pathological factors that lead to demineralization and protective factors that lead to remineralization. It is also considered that bacteria are an important etiologic factor leading to demineralization for secondary caries. Generally, the rationales of all the modification of restorative material or prevention of secondary caries normally include two fundamental points: one is the decrease of demineralization and/or increase of remineralization of the hard tooth tissues; the other is to interfere the metabolism of caries-related bacteria and/or to decrease the amount of bacteria/inhibit bacteria growth in the plaque or /and the carious dentin under restorations. Thus, in all the past years, most scientists and clinical dentists focused on adding anticaries

It has been well-known such restorative materials can release copper, Ag–Cu alloy, zinc, calcium, aluminum and uoride, which are able to inhibit bacteria growth or decrease colonization and acidogenicity of oral plaque, play antibacterial activities and reduce the rate of restoration replacement. The followings are several basic fillings used and researched by clinical dentists and scientists throughout the world: amalgam restorations, zinc oxide eugenol cement; common composite resin (CR); common glass ionomer cement (GIC); and

Clemens Boeckh et al. investigated the antimicrobial effects of five restorative materials and showed that the most remarkable inhibitory activity was observed with ZOE [Boeckh et al., 2002]. The antimicrobial effects of zinc oxide and ZOE are well recognized [Podbielski et al., 2000; Yap et al., 1999]. Unfortunately, ZOE cannot be widely used except for temporary filling due to its high solubility and insufficient mechanical properties. Different types of amalgam may have different effects on S. mutans growth and bacterial penetration [Fayyad and Ball, 1987]. For instance, a low-copper amalgam can decrease the lesion size significantly [Grossman and Matejka, 1995], non-gamma-2 amalgam can inhibit the metabolic activity of microorganisms due to the release of copper [Wallman-Björklund et al.,

It has been widely shown in long term studies that CR have higher rate of restorations replacement than GIC and amalgam [Leinfelder,et al., 1987; Collins et al., 1998]. An in situ study showed that the percentage of streptococci in plaque on different materials was to be 13.7% on composite, 4.3% on amalgam and 1.1% on glass ionomer cement [Svanberg et al., 1990]. Another study showed up to eight times more microbes beneath composite restorations compared to amalgam and suggested the type of restorative material may have influence on the composition of the microflora [Splieth et al., 2003]. The high rate replacement of CR occurred might be due to its shrinkage and non-fluoride release [Savarino et al., 2004]. Thus, scientists have been making great efforts to improve CR, such as reducing the polymerization shrinkage of composite, increasing the adhesion stress. Remarkably, a modified ion-releasing resin composite (IRCR) has been invented, which can

different ion-released restorative materials containing fluoride-containing materials.

**4. Prevention and treatment of secondary caries** 

**4.1 Prevention of secondary caries** 

substance into restorative materials.

1987].

smaller spaces, minerals dissolved from the tooth structure due to the acid attacks would supersaturate the space immediately and create the remineralization of tooth tissue and smaller wall lesions.

In contrast, other studies have not stated any association between gap presence and secondary caries [Kidd and O'Hara, 1990; Pimenta et al., 1995]. Some suggested that there was no caries lesion along the cavity wall, unless large voids or gaps of ≥ 250μm [Ozer and Thylstrup, 1995] or even ≥400μm [Kidd et al., 1995]. Jørgensen and Wakumoto found poor correlation between the two only when gaps was ≥50μm [Jørgensen and Wakumoto, 1968]. Thomas et al. found no clear caries along the cavity wall next to composite, but to acrylic resin through an in situ study [Thomas et al., 2007]. Besides, observed cracks in teeth might be the best clinical evidence for microleakage does not lead to dental caries. These cracks and the adjacent areas can be stained over time, however, there is no caries development. The stained component is considered to be the proteinaceous material in the crack or crevice, and similar in composition to that of the biofilm which normally covers all teeth and restorations [Mjör, 2005].

Recently, a few studies have been conducted concerning the relationship between gap size and secondary caries in the presence or absence of fluoride. Cenci et al. in 2008 suggested that microleakage did not seem to influence secondary caries while the presence of fluoride in the plaque like biofilm (PLB) provided either by glass ionomer cement (GIC) or fluoride dentifrice (FD) [Cenci et al., 2008]. In 2009, Cenci et al. demonstrated that carious lesion depth increased with gap size for composite resin (CR) and suggested that the gap width affected secondary caries formation at the cavity wall, but only in the absence of fluoride released from fluoride-containing materials, such as GIC [Cenci et al., 2009]. Thus, these findings give implications for clinical caries treatment choices and further studies in vivo or clinical experiments are needed to investigate the relationship between the gap size, fluoride presence and secondary caries.

In sum, up to the present, there is no specific conclusion on the relationship between microleakage and secondary caries, specially the wall lesion of the cavity. The possible reasons are as follows: 1) Oral cavity is such an extremely complex that it is impossible to simulate completely, so the research results may be not all-inclusive. 2) Secondary caries is caused by various factors, so it might be difficult for researchers to consider fully when designing their experiments. Thus, in consequence, different experiments bring out different results, and sometimes those results are even conflicting. 3) For individual differences, people have varying degrees of susceptibility to caries. Therefore, different clinical studies may lead to different results and conclusions. 4) Some clinical studies may lack reasonable designs, which lead to incomprehensive results. However, there is a consensus that microleakage is indeed associated with secondary caries due to the existence of bacteria. It seems that microleakage is just a necessary but not a sufficient condition for the formation of a wall lesion [Kidd et al., 1995; Thomas et al., 2007], although in 2009, Diercke et al. carried out a pure in vitro experiment, in which the development of the outer lesion in the secondary caries was inhibited to study the relation between the gap size and the wall lesion independently, and confirmed the occurrence of wall lesions without the presence of outer lesion and indicated the extent of wall lesion increased with increasing gap width ranging from 50 to 250μm [Diercke et al., 2009]. Therefore, more in-depth studies are needed to get a thorough understanding about the relationship between microleakage and secondary caries.

smaller spaces, minerals dissolved from the tooth structure due to the acid attacks would supersaturate the space immediately and create the remineralization of tooth tissue and

In contrast, other studies have not stated any association between gap presence and secondary caries [Kidd and O'Hara, 1990; Pimenta et al., 1995]. Some suggested that there was no caries lesion along the cavity wall, unless large voids or gaps of ≥ 250μm [Ozer and Thylstrup, 1995] or even ≥400μm [Kidd et al., 1995]. Jørgensen and Wakumoto found poor correlation between the two only when gaps was ≥50μm [Jørgensen and Wakumoto, 1968]. Thomas et al. found no clear caries along the cavity wall next to composite, but to acrylic resin through an in situ study [Thomas et al., 2007]. Besides, observed cracks in teeth might be the best clinical evidence for microleakage does not lead to dental caries. These cracks and the adjacent areas can be stained over time, however, there is no caries development. The stained component is considered to be the proteinaceous material in the crack or crevice, and similar in composition to that of the biofilm which normally covers all teeth and

Recently, a few studies have been conducted concerning the relationship between gap size and secondary caries in the presence or absence of fluoride. Cenci et al. in 2008 suggested that microleakage did not seem to influence secondary caries while the presence of fluoride in the plaque like biofilm (PLB) provided either by glass ionomer cement (GIC) or fluoride dentifrice (FD) [Cenci et al., 2008]. In 2009, Cenci et al. demonstrated that carious lesion depth increased with gap size for composite resin (CR) and suggested that the gap width affected secondary caries formation at the cavity wall, but only in the absence of fluoride released from fluoride-containing materials, such as GIC [Cenci et al., 2009]. Thus, these findings give implications for clinical caries treatment choices and further studies in vivo or clinical experiments are needed to investigate the relationship between the gap size, fluoride

In sum, up to the present, there is no specific conclusion on the relationship between microleakage and secondary caries, specially the wall lesion of the cavity. The possible reasons are as follows: 1) Oral cavity is such an extremely complex that it is impossible to simulate completely, so the research results may be not all-inclusive. 2) Secondary caries is caused by various factors, so it might be difficult for researchers to consider fully when designing their experiments. Thus, in consequence, different experiments bring out different results, and sometimes those results are even conflicting. 3) For individual differences, people have varying degrees of susceptibility to caries. Therefore, different clinical studies may lead to different results and conclusions. 4) Some clinical studies may lack reasonable designs, which lead to incomprehensive results. However, there is a consensus that microleakage is indeed associated with secondary caries due to the existence of bacteria. It seems that microleakage is just a necessary but not a sufficient condition for the formation of a wall lesion [Kidd et al., 1995; Thomas et al., 2007], although in 2009, Diercke et al. carried out a pure in vitro experiment, in which the development of the outer lesion in the secondary caries was inhibited to study the relation between the gap size and the wall lesion independently, and confirmed the occurrence of wall lesions without the presence of outer lesion and indicated the extent of wall lesion increased with increasing gap width ranging from 50 to 250μm [Diercke et al., 2009]. Therefore, more in-depth studies are needed to get a thorough understanding about the relationship between microleakage and

smaller wall lesions.

restorations [Mjör, 2005].

presence and secondary caries.

secondary caries.
