Diagnostic accuracy of kidney ultrasound compared to mercaptoacetyltriglycine-3 scan in paediatric multicystic dysplastic kidney disease.

BACKGROUND

Multicystic dysplastic kidney disease (MCDK) is a congenital kidney anomaly frequently misdiagnosed as hydronephrosis on antenatal ultrasound. Confirmatory nuclear imaging is seldom available in resource-limited settings. The study aimed to assess the diagnostic accuracy of kidney ultrasound (KUB scan) for detecting paediatric MCDK, using Mercaptoacetyltriglycine-3 scan ([99mTc]Tc-MAG3) differential renal function as the reference standard.

METHODS

A retrospective diagnostic accuracy analysis of consecutive children under 13 years with suspected unilateral MCDK, who underwent both KUB and [99mTc]Tc-MAG3 scans within 6-8 weeks of presentation at Red Cross War Memorial Children's Hospital between January 2014 and December 2023 was done. Diagnosis required characteristic MCDK features on KUB and absent function on [99mTc]Tc-MAG3 scans. Reporting adhered to the STARD guidelines.

RESULTS

Of 793 eligible children, the [99mTc]Tc-MAG3 classified 101/101 (100.0%) kidneys as non-functional. The KUB scan accurately identified 97/98 and demonstrated a sensitivity of 99.0% (95% CI: 95.6-99.9%), specificity of 99.9% (95% CI: 99.4-100.0%), PPV of 99.0% (CI: 95.6-99.9%), NPV of 99.9% (99.4-100.0%) and overall accuracy of 99.7% in comparison to [99mTc]Tc-MAG3. Cohen's kappa indicated substantial agreement (0.988), with an AUC of 0.995 (95% CI: 0.983-1.000). Three missing KUB reports were excluded; sensitivity analysis accounting for missing data did not alter these findings.

CONCLUSIONS

Postnatal kidney ultrasound scan performs excellently in diagnosing paediatric MCDK in the South African context and obviates the need for confirmatory nuclear imaging. This has clinical practice implications for paediatric nephro-urology and supports its use as a primary diagnostic tool both in resource-limited and resource-sufficient settings.