Methods
With the aim to identify all publications in which a HPV vaccination was given in addition to the standard treatment for cervical dysplasia, we conducted an extensive search with a medical librarian of the literature from inception till 1th of august 2019 using the following databases: Pubmed, Medline and the Cochrane library. The search strategies contained various combinations of keywords and MeSH headings: Papillomavirus vaccines, HPV vaccine, human papilloma virus vaccination, Papillomaviridae, HPV, LEEP, conization, CIN, surgical therapy and biopsy. The search strategy is presented in appendix S1.
Eligible studies were human studies published in English or Dutch, without any restriction on study design. All articles that described the effect of vaccination in addition to a surgical treatment such as LEEP, conization, cone biopsy or other surgical therapy for cervical intra-epithelial neoplasia (CIN I-II-III) or high grade squamous intraepithelial lesion (HSIL) could be included. No limitation were made on age, number of sexual partners, HPV-type or prior treatment for HPV related diseases. Studies on HIV were excluded.
The titles and abstracts of all the citations retrieved by the searches were screened for relevance independently and separately by two of the authors (RL and WH). The full texts of the resulting articles were assessed on their relevance to this review. In case of disagreement, the third author (HB) assessed the relevance as well. The references of selected publications were checked on relevant other publications. The following details were sourced: Author, year of publication, age, study type, sample size, recurrence rate of cervical dysplasia after treatment and vaccination with a HPV vaccine. A random effect model is used to calculate the pooled effect size, anticipating heterogeneity between studies. Categorical data were summarized using the Mantel-Haenszel Risk Ratio (RR) with 95% confidence intervals (CI). Heterogeneity was calculated using the I² statistic. Meta-analysis was performed using the software “Review Manager” version 5.3 from the Cochrane Collaboration.
Risk of bias was assessed using the modified Downs and Black (D&B) scale. The Downs and Black scale is also suitable for non-randomised trials (20). We used the checklist to score the reporting, the internal validity and external validity of the eligible studies. A High score indicates a better methodological quality. A score of 19 or higher indicates a high quality of the study as used in other studies (21). Whereas a score of <19 considered to be of low quality (See online supplementary table S2). The quality assessments of the included studies were used to categorize the level of evidence. We scored all studies below 19 and were subsequently methodologic of low quality.