Association between passive and active smoking evaluated by salivary cotinine and periodontitis

Abstract

These findings suggest that passive smoking classified in terms of salivary cotinine level may be an independent periodontitis risk indicator.

Systematic Review and Meta-analysis of the Association Between Exposure to Environmental Tobacco Smoke and Periodontitis Endpoints Among Nonsmokers

“…Indeed, among studies that reported both adjusted and unadjusted estimates, [20][21][22][23] our assessment of the validity-precision trade-off 24 favored adjustment in all except one (see Supplementary Table S4). 20 Nonetheless, we used unadjusted estimates when they were the only estimates available, calculating them ourselves when necessary. Therefore, we assessed adjusted (ie, author reported) versus unadjusted (calculated by us) POR estimates as a study characteristic.…”

“…From 10 publications [14][15][16][17][20][21][22][23]25,29 and two conference abstracts 30,31 systematically reviewed, eight [16][17][18][19][20][21][22][23]25,31 were eligible for meta-analysis. Reports were excluded for two main reasons, and two studies 18,19 were ineligible for both reasons.…”

“…First, we excluded three reports 18,19,30 that were conducted in the same sample as studies already included. 20,22 Second, we excluded three reports that could not be placed on the same quantitative footing as the rest, that is, we could not obtain POR for CAL and/or PD or tooth loss because these studies either reported mean (SE) levels of CAL and/or PD, 29 odds ratio for change in CAL from baseline, 19 or mean (SE) for salivary biomarkers related to periodontitis but not on actual measures of CAL and/or PD. 18 Funnel plots of ln(POR) versus SE (ln POR) were visually examined for asymmetry and quantitatively assessed using the rank correlation test of Begg and Mazumdar, 32 the regression-based test of Egger et al, 33 and the Duval and Tweedie 34 trim-and-fill imputation method.…”

“…The limited number of studies precluded us from conducting a multiple meta-regression, therefore we assessed each study characteristic one at a time. Study characteristics analysis was conducted only for studies reporting CAL and/or PD (n = 6) [20][21][22][23]25,31 and not for those reporting tooth loss (n = 2) because of the too few number of studies.…”

“…[41][42][43][44] Of the three conference abstracts reviewed, 30,31,45 one was excluded because of incomplete data reporting. 45 In total, 10 observational full-text articles [16][17][18][19][20][21][22][23]25,29 and two conference abstracts 30,31 were systematically reviewed and are presented in Table 1 with a detailed description in Supplementary Table S2. Eight of these were eligible for meta-analysis, six reporting measures of CAL and/or PD were meta-analyzed separately from the two reporting tooth loss.…”