Percentage agreements for MRI scans based on a margin of error were compared to alternative tests in six studies (Supplementary Information Resource 4). The six MRI and US studies (Balu-Maestro et al, 2002; Julius et al., 2005; Yeh et al, 2005; Akazawa et al, 2006; Segara et al, 2007; Guarneri et al, 2011); MRI was compared with clinical examination in four studies (Balu-Maestro et al. Yeh et al, 2005; Akazawa et al, 2006; Segara et al. 2007) and mammography in three studies (Balu-Maestro et al, 2002; Julius et al., 2005; Yeh et al., 2005). For all but one study and beyond the range of reported margins of error, estimates of percentage match for MRI were higher than for comparative tests. With an exception to this results model, a study showing several margins of error (Segara et al., 2007) found a higher percentage of agreement for MRI than for the United States for ±0 and ±1 cm margins, but the percentage agreeing with ±2 cm was slightly higher for the United States (92%) MRI (88%). In another study (Guarneri et al., 2011), the percentage difference in the MRI agreement was relatively small compared to the United States (20% vs. 15% for ±0 cm; 54% vs. 51% for ±0.5 cm and 71% versus 68% for ±1 cm). Secondary analysis assessed the relative correspondence between the number of case reports showing a change in a given outcome and the SOE score (1 considered insufficient; 2 if low; 3 as moderate and 4 as high) reported in the meta-analysis of the clinical trial. The spearman rank correlation was used to assess relative compliance. The results were presented in a dispersal diagram. Similarly, the relative match rate between the quantitative data score, based on case reports with the modified ERT results, and the SOE was good (Rho – 0.82, 95%CI: 0.43 to 0.95) when the strong confirmation method was used (Fig.
2). Conversely, analysis of the modified ERT results in case reports on the basis of the low confirmation method revealed a moderate approval rate (Rho – 0.63, 95%IC: 0.044 to 0.89) with the SOE (see additional file 1: Figure S1). Our proposal provided a meta-analysis of cases of patients with MPS II treated with ERT, comparing the degree of evidence assigned to each result with what was attributed in a previous meta-analysis clinical trial published by an independent research group. In a population with MPS-II, we tried to confirm the impressive consistency between case reports and meta-analyses of clinical trials in patients with MPS-I . Decision structure to assess when AD meta-analysis HRs are likely to be reliable and when the IPD approach might be needed.