Selection of caries-risk patients
Inability of a sole salivary MS test to predict caries risk
As already mentioned in this chapter, numerous cross-sectional as well as longitudinal
studies have shown significant correlations between salivary MS levels and caries
prevalence and caries incidence (for review, see Bratthall, 1991; Bratthall and
Ericsson, 1994; Beighton et al, 1989). At the surface level, even more significant
correlations between MS colonization and caries incidence have been found
(Axelsson et al, 1987b; Kristoffersson et al, 1985).
Most of the early salivary MS studies were carried out in child populations with
relatively high caries prevalence (Sweden in the 1970s), and at that time more than 1
million CFUs of MS/mL of saliva was shown to be a good predictor of caries risk
(Klock and Krasse, 1977; Zickert et al, 1982). However, since then, caries prevalence
in Sweden and many other industrialized countries has decreased significantly. The
correlation between one single etiologic factor, such as salivary MS levels, and caries
prevalence and caries incidence tends to be weaker in such populations, because
dental caries is a multifactorial disease.
In a more recent 2-year longitudinal Swedish study in children (5 to 7 years and 12 to
14 years), Sullivan et al (1989) found that the correlation between caries incidence
and both salivary MS and lactobacilli was weak at the individual level, particularly
after correction for confounding factors, such as oral hygiene status. In another study,
Sullivan et al (1996) found that MS and lactobacilli, whether in saliva or in plaque,
was not a powerful enough tool for caries prediction in a group of 14 to to 15 year
olds. Kingman et al (1988a) also found that the predictive values for salivary MS and
lactobacilli on caries incidence in 10- to 15-year-old US schoolchildren was low (31%
and 39%, respectively). The moderate-to-low predictive value of salivary MS may
partly be explained by differences in virulence, not only among species of MS but
also among individual clones of S mutans and S sobrinus (Bowden, 1997). Even in
relation to root caries, the important role of MS and lactobacilli has recently been
questioned (Schupbach et al, 1995; for recent reviews on the importance of the
specific microflora for prediction of caries risk, see Bowden, 1997; Bowden and
Edwardsson, 1994; Bratthall and Ericsson, 1994; Hausen et al, 1994; Hausen, 1997.
More recent cross-sectional studies in Swedish schoolchildren have repeatedly found
that the cutoff for correlation between salivary MS counts and caries prevalence is
MS negative or MS positive rather than > 1 million CFUs of MS/mL of saliva (see
Fig 23) (Kristoffersson et al, 1986). However, the dilemma is that only 10% to 30% of
the individuals are MS negative in most populations (higher in young children and
lower in elderly). The question is how to select 5% to 25% high- and very high-caries
risk individuals from among the 70% to 90% MS-positive subjects.
