TABLE 1.
Summary table of systematic reviews with meta‐analysis.
| Systematic review | Population | No. of studies | No. of participants | Index prognostic factor (cervical length) | Outcome definition (spontaneous preterm birth, < gestational age in weeks) | Timing (GA at measurement) | Methods | |||
|---|---|---|---|---|---|---|---|---|---|---|
| Cervical length cutoff values (mm) | No. of studies per cutoff set | Design | Data source | Data synthesis | ||||||
| Leitich 1999 | Singleton, normal risk a | 6 | 106–3694 | 20, 25, 29, 30, 34, 35, 39, 40, 45, 50, 55 | 1–2 | 35, 37 | 20, means: 20+4, 23+4, 27+6; 29 | Diagnostic | Aggregate | Study sensitivities/ specificities plotted as ROC curve |
| Twin, unselected | 1 | 253 | 25 | 1 | 32, 35, 37 | 23, 27.5 (mean) | ||||
| Honest 2003 | Singleton, normal or high‐risk a | 18 | 50–6877 | 10, 15, 20, 22, 27, 29, 30, 35, 39, 40, 45, 50, 60 | 1–5 | 32, 34, 37 | <20, 20–24, >24 | Diagnostic | Aggregate | sROC curves, LR+/LR− |
| Twin, unselected or high‐risk | 11 | 43–434 | 15, 20, 25, 30, 35, 40, 45 | 1–4 | 32, 34, 37 | <20, 20–24, >24 | ||||
| Krupa 2006 | Singleton and twin | 6 | 32–2915 | Not described | Not described | Not described | Not described | n/a | n/a | n/a |
| Crane 2008 | Singleton, high‐risk | 14 | 42–1476 | 10, 15, 20, 25, 30, 35, 40, 45 | 1–9 | 26, 28, 30, 32, 34, 35, 37 | <20, 12–24, 14–18, 14–24, 15–20, 15–24, 16–19, 18–22, 20–24, <24, 24–28, 24–30, 25–29, <32 | Diagnostic | Aggregate | Fixed effects meta‐analysis, summary LR+/LR− |
| Honest 2009 | Singleton, normal or high‐risk a | 13 | 69–6334 | 15, 20, 22, 25, 27, 30, 35, 39 | 1–2 | 34, 37 | <20, 20–24, >24 | Diagnostic | Aggregate | Meta‐analysis technique not further described. Summary LR+/LR− |
| Conde‐Agudelo 2010 | Twin, unselected (one study post‐selective reduction of triplets) a | 16 | 18–1135 | 15–20, 15.9 ± 0.3, 18, 18–24, 18–26, 18–36, 20, 20–25, 20–32, 21–23, 22–24, 23, 23–33, 24, 24–26, 24–36, 25–28, 25–32, 26–28, 28, 32, <34, 38 | 2–6 | 34, 37 | 20–24 | Diagnostic | Aggregate | Bivariate random‐effects meta‐analysis. sROC curves, sensitivity, specificity, LR+/LR− |
| Domin 2010 | Singleton, normal or high‐risk | 23 | 57–6877 | 20, 25, 33.15 | ns | 35, 37 | <20, >20, 14–24 | Diagnostic | Aggregate | Random‐effects meta‐analysis, ROC curves. AUC, sensitivity, specificity, LR+/− |
| Lim 2011 | Twin, with or without additional risk factors | 16 | 14–434 | 20, 25, 30, 35 | not reported | <30, 30–34, <34 | <20, 20–24, >24 | Diagnostic | Aggregate | Bivariate regression analysis. Sensitivity, specificity |
| Honest 2012 (data from 2009) | Singleton, normal or high‐risk a | ns | ns | 15, 20, 22, 25, 30, 32.5 | 1–5 | 34, 37 | 14–20, 20–24 | Diagnostic | Aggregate | Meta‐analysis technique not described. Summary LR+/− |
| Kleinrouweler 2013 | Singleton, previous SPTB a | 7 | 37−76 476 | 15, 15–20, 20–25, 25–30, ≥30 | ns | 32, 34, 37 | <15, 15–20, 20–25, 25–30, ≥30 | Prognostic | Individual participant | Univariable analysis, wo‐level random intercept logistic regression models, Cox regression analysis, % risk |
| Conde‐Agudelo 2014 | Twin, unselected (one study post‐selective reduction of triplets) | 16 (single measure) | 434–1955 | 20, 25 | 3–5 | 28, 32, 34, 37 | 20–24 | Diagnostic | Aggregate | Meta‐analysis technique not described. ROC curve or sensitivity, specificity, LR+ |
| 7 (shortening over time) | 1004 | Any shortening | ns | 28–36 | ns | |||||
| Barros Silva 2014 | Singleton (normal or high‐risk) or twin | 21 | 80–5068 | 10, 15, 20, 22, 23.3, 25, 26, 28, 29, 30, 35 | 1–8 | 26, 28, 30, 32, 33, 34, 35, 37 | 18, 18–21, 18–22, 18–24, 20–24, 21–24, 22–24, 23, 24 | n/a | n/a | n/a |
| Conde‐Agudelo 2015 | Singleton, normal or high‐risk | 6 | 68–2531 | Any shortening | 1–5 | 35, 37 | Disregarded | Diagnostic | Aggregate | Bivariate random effects meta‐analysis, sROC curves. Sensitivity, specificity, LR+ |
| Twin, unselected | 7 | 20–209 | Any shortening | 3–7 | 28, 30, 32, 34 | |||||
| Kindinger 2016 | Twin, without additional risk factors a | 12 | 22–1138 | 5, 10, 15, 20, 25, 30, 35, 40 | 12 | 28, 28–32, 32–36, ≥36 | 18–26 | Prognostic | Individual participant | Multinomial logistic regression, multinomial log‐linear model. % risk |
Note: Where two lines of data are listed for a single review, the first line indicates the thresholds reported by the review. The second line indicates the thresholds reported in the included primary studies.
Abbreviations: AUC, area under the curve; GA, gestational age; LR, likelihood ratio; n/a, not applicable; ns, not specified; ROC, receiver operating characteristic; sROC, summary receiver operating characteristic.
a
Includes unpublished data.