The diagnostic accuracy of ultrasound and genomic tests for the diagnosis of autosomal-dominant polycystic kidney disease: a systematic mapping review

Abstract

Background: Genomic and ultrasound tests can provide diagnostic and prognostic information on autosomal-dominant polycystic kidney disease (ADPKD), and can screen first-degree relatives in whom early diagnosis can be advantageous. We conducted a systematic mapping review on test accuracy and characteristics over time.

Methods: Medline, Embase, and Cochrane were searched (August 2023) for studies in first-degree relatives/individuals clinically diagnosed with ADPKD receiving genomic or ultrasound tests. Acceptable reference standards for sensitivity/detection rate and specificity were definitive imaging or genomic confirmation. Genomic studies were categorized by technology and read length. Relationships between sensitivity, specificity, genomic technology, diagnostic criteria/reference standard, and genes tested were compared.

Results: From 1029 non-duplicate titles retrieved, 51 genomic and 7 ultrasound studies were included. There were no genomic studies in first-degree relatives. Among studies in patients with clinical diagnoses, genomic sequencing methodologies were highly heterogeneous [next generation (short read (n = 20), long read (n = 1)), targeted Sanger (n = 19), whole exome (n = 1) with additional multi-ligation probe analysis (n = 13)]. Median sensitivity was 78% (Interquartile range 65% to 88%). Ultrasound sensitivity and specificity generally improved with age and were worse in PKD2 patients compared to PKD1 (lowest reported 31% and 88%, respectively, in polycystic kidney disease (PKD) 2 patients aged 5-14; highest 100% and 100%, respectively, in multiple gene/age categories).

Conclusions: Despite technological advances, sensitivity of genomic tests appeared static between 2000 and 2023. Possible explanations include clinical diagnostic criteria (and hence populations recruited) widening from PKD1 to include PKD2 and atypical phenotypes, and small incremental gains of testing genes other than PKD1 and PKD2. For people at risk of ADPKD in genetically unresolved families, the accuracy of ultrasound is uncertain. Unified genomic test taxonomies would facilitate future reviews. Registration: PROSPERO CRD42023456727.

Keywords: ADPKD; detection rate; diagnosis; genomic tests; ultrasonography.

Graphical Abstract

Figure 1:

PRISMA flow chart showing the process of study selection for the review.

Figure 2:

Map of origin of included studies.

Figure 3:

Charts of study characteristics over time. (a) Diagnostic criteria/radiological reference standard for inclusion of individuals with clinical diagnosis of ADPKD by year of study publication; (b) genomic test technology by year of study publication; and (c) genes analysed by genomic tests by year of study publication.

Figure 4:

Diagnostic test accuracy (proportion with genomic variants classed as definitely pathogenic, pathogenic, likely pathogenic, and probability pathogenic or similar terms), stratified by genes targeted and recruitment criteria, by study publication year. Blue line, studies recruiting according to Ravine [28], with genomic testing for PKD1 or more; green line, studies recruiting according to Pei 2009/2015 [18, 19], with genomic testing for PKD1 and PKD2 or more; red line, studies recruiting according to other criteria, with genomic testing for more than PKD1 and PKD2. Studies weighted by size when estimating longitudinal changes and 95% confidence intervals (grey).

Figure 5:

Sensitivity and specificity of ultrasound studies. Each bar represents the sensitivity or specificity for the age range spanned by the bar, as reported by individual studies included in this review.