The association between second-hand smoke and the periodontal status of children and adolescents

Abstract

Data sources The authors searched Embase, Medline/PubMed, Web of Science, Scopus and LILACS ranging from inception to 23 November 2021. A grey literature search was run utilising OpenGrey along with Google Scholar (most relevant first 200 hits). For the identification of additional articles of relevance, a hand-search was also conducted upon the list of references of all the included studies. No restrictions in terms of date and language were utilised in the search. Controlled vocabulary in the form of free text and MeSH terms was utilised to execute the search strategy.

Study selection A screening of the studies potentially satisfying the eligibility criteria was conducted independently by two reviewers on the basis of their abstracts and titles. The records so retrieved were classified under the headings of 'exclude', 'include' and 'uncertain'. Upon the completion of this exercise, an acquisition was made of the full texts of papers judged relevant. Data extraction was performed on all the studies which met the eligibility criteria. Diverging views among the reviewers were resolved with consensus. The judgement of a third reviewer was sought to resolve any disagreements.

Data extraction and synthesis Information extraction covered measurement details (cotinine levels of children and/or self-reported questionnaires from children and/or parents); study identification (year of publication, first author and location); distribution according to the status of second-hand smoke and sample characteristics; exposure criteria (according to cotinine levels or severity of parental smoking); analytic approach utilised (multivariate or bivariate analysis); results along with assessment of outcomes (partial- or full-mouth protocol used) and criteria (index type used). Due to only cross-sectional studies being included upon retrieval, the methodological quality assessment was performed utilising an adapted variant of the Newcastle-Ottawa scale. The judgement criteria used included non-response rate, sample size and representativity, confounding control, determination of exposure, outcome assessment and statistical analysis. Evidence tables were built utilising qualitative synthesis in order to ascertain differences in the results and characteristics of the included studies. In case the eligibility criteria were met by three or more papers, the data was subjected to pair-wise comparisons between those unexposed to second-hand smoke with those exposed to it in terms of the specified periodontal outcome measures. Periodontal outcomes were evaluated as continuous measures with the random-effects model. Cochran's Q-test was utilised to assess the heterogeneity and in cases where the heterogeneity was unexplained, a sensitivity analysis was conducted to evaluate whether a single study exerted a significant effect upon the pooled results.

Results Six hundred and thirty-seven records were obtained from the search. After screening of the abstracts and titles, eight studies met the eligibility criteria subsequent to the reading of the complete text of 16 reports. One study was conducted in a high socio-demographic index (SDI) country with four in the middle-high, one in the middle and two in the lower-middle SDI countries. Four studies did not provide details pertaining to outcome assessment, while one included a full-mouth and another three, a partial-mouth assessment protocol. Two studies were deemed to be of high quality in terms of methodological quality assessment. Three reports possessed a satisfactory sample size while none of the included studies included data relating to non-response rate. The authors deemed meta-analysis to be possible only in the case of periodontal probing depth (PPD) and gingival index (GI) scores. There was a positive association reported between a high GI score and second-hand smoke exposure. There was, however, no difference present at the subgroup level in terms of the SDI of the country of the study, the age of the cohort as well as the period in which the study was carried out. In contrast, a significant effect was observed for the location of the study data collection in that children from areas with a higher current-smoker prevalence exhibited greater GI scores as a result of increased second-hand smoke exposure. There was no observable difference from the pooled estimate for second-hand smoke exposure compared to the unexposed cohort in terms of PPD. No effects were demonstrable at the subgroup level upon data stratification for the study country SDI and period of data collection. There was greater clinical attachment loss among children who were exposed to second-hand smoke, particularly for those whose fathers were smokers despite adjustments being made for gender, plaque scores and age.

Conclusions The systematic review demonstrates a very low level of certainty in evidence to point towards second-hand smoke exposure of children up to 15 years of age leads to higher GI scores as compared to their unexposed counterparts. Even though no differences have been demonstrated in terms of PPD, there is a necessity for further research to draw conclusions on other pertinent periodontal parameters such as bleeding on probing and clinical attachment loss.