The rate of missed diagnosis of lower-limb DVT by ultrasound amounts to 50% or so in patients without symptoms of DVT: A meta-analysis: Erratum

In the article, “The rate of missed diagnosis of lower-limb DVT by ultrasound amounts to 50% or so in patients without symptoms of DVT: A meta-analysis”,^[1] which appears in Volume 98, Issue 37 of Medicine, the authors need to correct Figure 7 and the related discussion in section 3.7. The corrected section and figure are below.

Figure 7.

The subgroup analyses of 2000s, 1990 s and 1980 s for DVT.

3.7. The subgroup analyses of 2000s, 1990 s and 1980 s for DVT

Seven individual studies of 2000 s that provided available DATA were included in the quantitative meta-analysis. Subgroup analysis showed that 2000 s for DVT had a poor sensitivity of 42% (95% CI = 37–48%; Fig. 7A) with significant heterogeneity (I² = 94.0%, p = 0.00), and a higher specificity of 95% (95% CI = 94–96%; Fig. 7B) with significant heterogeneity (I² = 92.1%, p = 0.00).

Thirty-one individual studies of 1990 s that provided available DATA were included in the quantitative meta-analysis. Subgroup analysis showed that 1990 s for DVT had a moderate sensitivity of 58% (95% CI = 53–62%); Fig. 7C) with heterogeneity (I² = 83.7%, p = 0.00), and a higher specificity of 97% (95% CI = 97–98%; Fig. 7D) with significant heterogeneity (I² = 80.6%, p = 0.00).

Three individual studies of 1980 s that provided available DATA were included in the quantitative meta-analysis. Subgroup analysis showed that 1980 s for DVT had a moderate sensitivity of 0.58% (95% CI = 53–62%; Fig. 7E) with heterogeneity (I² = 83.7%, p = 0.00), and a higher specificity of 97% (95% CI = 95–99%; Fig. 7F) with heterogeneity (I² = 0.0%, p = 0.98).