Health Technology Assessment (Oct 2017)
Spot protein–creatinine ratio and spot albumin–creatinine ratio in the assessment of pre-eclampsia: a diagnostic accuracy study with decision-analytic model-based economic evaluation and acceptability analysis
Abstract
Background: The National Institute for Health and Care Excellence (NICE) guidelines highlighted the need for ‘large, high-quality prospective studies comparing the various methods of measuring proteinuria in women with new-onset hypertensive disorders during pregnancy’. Objectives: The primary objective was to evaluate quantitative assessments of spot protein–creatinine ratio (SPCR) and spot albumin–creatinine ratio (SACR) in predicting severe pre-eclampsia (PE) compared with 24-hour urine protein measurement. The secondary objectives were to investigate interlaboratory assay variation, to evaluate SPCR and SACR thresholds in predicting adverse maternal and fetal outcomes and to assess the cost-effectiveness of these models. Design: This was a prospective diagnostic accuracy cohort study, with decision-analytic modelling and a cost-effectiveness analysis. Setting: The setting was 36 obstetric units in England, UK. Participants: Pregnant women (aged ≥ 16 years), who were at > 20 weeks’ gestation with confirmed gestational hypertension and trace or more proteinuria on an automated dipstick urinalysis. Interventions: Women provided a spot urine sample for protein analysis (the recruitment sample) and were asked to collect a 24-hour urine sample, which was stored for secondary analysis. A further spot sample of urine was taken immediately before delivery. Main outcome measures: Outcome data were collected from hospital records. There were four index tests on a spot sample of urine: (1) SPCR test (conducted at the local laboratory); (2) SPCR test [conducted at the central laboratory using the benzethonium chloride (BZC) assay]; (3) SPCR test [conducted at the central laboratory using the pyrogallol red (PGR) assay]; and (4) SACR test (conducted at the central laboratory using an automated chemistry analyser). The comparator tests on 24-hour urine collection were a central test using the BZC assay and a central test using the PGR assay. The primary reference standard was the NICE definition of severe PE. Secondary reference standards were a clinician diagnosis of severe PE, which is defined as treatment with magnesium sulphate or with severe PE protocol; adverse perinatal outcome; one or more of perinatal or infant mortality, bronchopulmonary dysplasia, necrotising enterocolitis or grade III/IV intraventricular haemorrhage; and economic cost and outcomes. Health service data on service use and costs followed published economic models. Results: In total, 959 women were available for primary analysis and 417 of them had severe PE. The diagnostic accuracy of the four assays on spot urine samples against the reference standards was similar. The three SPCR tests had sensitivities in excess of 90% at prespecified thresholds, with poor specificities and negative likelihood ratios of ≥ 0.1. The SACR test had a significantly higher sensitivity of 99% (confidence interval 98% to 100%) and lower specificity. Receiver operating characteristic (ROC) curves were similar (area under ROC curve between 0.87 and 0.89); the area under the central laboratory’s SACR curve was significantly higher (p = 0.004). The central laboratory’s SACR test was the most cost-effective option, generating an additional 0.03 quality-adjusted life-years at an additional cost of £45.07 compared with the local laboratory’s SPCR test. The probabilistic analysis showed it to have a 100% probability of being cost-effective at the standard willingness-to-pay threshold recommended by NICE. Limitations: Implementation of NICE guidelines has led to an increased intervention rate in the study population that affected recruitment rates and led to revised sample size calculations. Conclusions: Evidence from this clinical study does not support the recommendation of 24-hour urine sample collection in hypertensive pregnant women. The SACR test had better diagnostic performance when predicting severe pre-eclampsia. All four tests could potentially be used as rule-out tests for the NICE definition of severe PE. Future work: Testing SACR at a threshold of 8 mg/mmol should be studied as a ‘rule-out’ test of proteinuria. Trial registration: Current Controlled Trials ISRCTN82607486. Funding: This project was funded by the National Institute Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 21, No. 61. See the NIHR Journals Library website for further project information.
Keywords
