Table 7.
Snuff use and endpoints relevant to periodontal and gingival disease – evidence from Sweden
| Study | Endpoint | Time of exposure | Statistic | Exposeda | Unexposeda | Nb | Summary of statistical testsc |
| Wickholm 2004 [10] | Plaque index | Ever | % ≥ 2.0 | 2.6 | 2.0 | 4 | OR 1.29 (0.45–3.70)d |
| Montén 2006 [12] | Plaque | Current | % | 59.0 | 64.0 | 19 | OR 0.75 (0.32–1.76) |
| Wickholm 2004 [10] | Calculus index | Ever | % ≥ 2.0 | 5.9 | 3.8 | 9 | OR 1.57 (0.76–3.23)d |
| Montén 2006 [12] | Gingivitis | Current | % | 47.0 | 50.0e | 16 | OR 0.94 (0.41–2.15) |
| Modeer 1980 [6] | Gingival index | Not known | mean | 1.10 | 0.89 | - | p < 0.001f |
| Rolandsson 2005 [20] | Gingival index | Current | mean | 12.4 | 13.1 | - | Difference not significant |
| Wickhol m 2004 [10] | Gingival index | Ever | % ≥ 2.0 | 8.5 | 12.0 | 13 | OR 0.68 (0.38–1.22)d |
| Rolandsson 2005 [20] | Gingival bleeding | Current | % | 10.0 | 20.0 | 4 | OR 0.44 (0.12–1.62) |
| Bergström 2006 [11] | Gingival bleeding | Current | - | - | -e | - | No significant difference |
| Former | - | - | -e | - | No significant difference | ||
| Frithiof 1983 [17] | Gingival recession | Current | % | 9.5 | - | 2 | - |
| Andersson 1989 [18] | Gingival recession | Currentg | % | 17.8 | - | 44 | Prevalence higher in loose snuff users (23.5%) than in portion-bag users (2.9%), p < 0.05 |
| Wickholm 2004 [10] | Gingival recession | Ever | % | 64.7 | 59.9 | 99 | OR 1.22 (0.87–1.73)d |
| Rolandsson 2005 [20] | Gingival recession | Current | % | 17.5 | - | - | - |
| Montén 2006 [12] | Gingival recession | Current | % | 42.0 | 17.0 | 14 | OR 3.72 (1.40–9.99)h |
| Wickholm 2004 [10] | Pocket depth | Ever | %≥ 5 mm | 10.5 | 9.5 | 16 | OR 1.11 (0.64–1.92)d |
| Bergström 2006 [11] | Pocket depth | Current | - | - | -e | - | No significant difference |
| Former | - | - | -e | - | No significant difference | ||
| Montén 2006 [12] | Pocket depth | Currentg | mean | 2.3 | 2.4 | - | No significant difference |
| Montén 2006 [12] | Attachment loss | Currentg | mean | 0.2 | 0.1 | - | No significant difference |
| Montén 2006 [12] | Alveolar bone level | Currentg | mean | 1.3 | 1.3 | - | No significant difference |
| Bergström 2006 [11] | Bone Heightj | Current | mean | 1.0 | 1.06e | - | No significant differencei |
| Former | mean | 1.12 | 1.06e | - | No significant differencei | ||
| Wickholm 2004 [10] | Periodontal diseasel | Current | - | - | - | - | OR 0.66 (0.30–1.32)k |
| Former | - | - | - | - | OR 2.55 (0.80–6.80)k |
a Exposure is always to snuff and is classified, where possible, as current or former. The corresponding unexposed group is never for ever, and non-current for current, except where indicated
b Number of exposed subjects with endpoint (where available)
c Tests are unadjusted for any potential confounding variable, except where stated. Where necessary ORs and 95% CIs are calculated from the data provided in the source paper
d The source paper presented results separately for four groups: A = never used tobacco, B = smoked only, C = snuff only, D = smoked and snuff. The ORs given in Table 7 are based on combining ORs for nonsmokers (C vs A) and smokers (D vs B) using fixed-effects meta-analysis [87], and are thus adjusted for smoking. ORs (CIs) specifically for nonsmokers are plaque index 1.13 (0.14–9.11), calculus index 3.53 (0.93–13.45), gingival index 1.14 (0.39–3.33), gingival recessions 1.43 (0.80–2.55) and pocket depth 1.61 (0.54–4.80)
e Unexposed is never snuff
f Adjusted for plaque index
g Snuff only, no smoking
h The ORs and CIs are adjusted for plaque, gingivitis and toothbrushing
i Adjusted for age
j Distance from the cement-enamel junction to the periodontal bone crest
k Adjusted for smoking and plaque index
l Three or more teeth with pocket depth ≥ 5 mm