الفهرس 
Pathgenesis of preeclampsiaOnly 14 pages are availabe for public view
 |  | from | 140 |  |  |
| | | from | 140 | | |
Abstract
In women with hypertensive disorders with pregnancy, the gold standard for diagnosis of significant proteinuria is 24-hour urine collection. However, 24-hour urine collection for measuring total proteinuria is time-consuming, subject to collection error, requires good patient’s compliance and usually done on an in-patient basis. The rationale for necessity of collection of urine for 24 hours for measuring total protein excretion is the proven variation in protein excretion throughout the day.
The synchronous excretion of creatinine and protein is reasonably constant throughout the day when the glomerular filtration rate is stable. Accordingly, some have proposed the use of protein to creatinine ratio in urine samples collected over shorter time periods, or even random urine samples, in order to overcome the problem of variation of protein excretion throughout the day.
The aim of the current work was to evaluate the reliability of PCR in both random and 4-hour urine samples as predictors of significant proteinuria in women with hypertensive disorders with pregnancy.
This observational cross-sectional study was carried out at Ain Shams University Maternity Hospital during the period from August 2009 till December 2009. The study included women with any form of hypertensive disorder with pregnancy after 20 weeks’ gestation, diagnosed as: a systolic arterial blood pressure ≥ 140 mm Hg, and/or a diastolic arterial blood pressure ≥ 90 mm Hg, with or without proteinuria (detected by dipsticks). Women with chronic renal disease, chronic liver disease, current or recent urinary tract infection or chronic hypertension with pre-existing proteinuria were excluded. Random urine sample was taken for checking proteinuria by dipsticks. A random urine sample was taken for checking both protein and creatinine levels, and the urinary protein-to-creatinine ratio (PCR) was calculated.
24-hour urine was collected for measuring total protein. The voided urine in the first 4 hours was collected in a separate container. A sample was taken from collected urine after the first 4 hours for checking the urinary PCR. Then the random urinary protein-creatinine ratio and first 4-hour urinary protein-creatinine ratio were calculated. The total 24 hour urine protein (mg) as the gold standard was measured as the summation of the first 4-hour and consecutive 20-hour urine protein.
A total 100 pregnant women were included in analysis. All included women had one form of hypertensive disorder with pregnancy (whether gestational hypertension, mild pre-eclampsia or severe pre-eclampsia). Of the included 100 women, 48 (48%) women had a non-proteinuric gestational hypertension, 46 (46%) women had mild pre-eclampsia and 6 (6%) women had severe pre-eclampsia. The mean total 24-hour urine protein was 769.64 mg per 24 hours (± 959.29, range: 110-5000). The mean urine PCR in random sample was 0.3 (± 0.12, range: 0.02-2.9). The mean urine PCR in 4-hour samples was 0.42 (± 0.21, range: 0.05-3.1). Our study showed a significant positive correlation between total 24-hour urinary protein and PCR in both random (r = 0.703, p < 0.001) and 4-hour (r = 0.651, p < 0.001) urine samples. There was a significant positive correlation between total 24-hour urinary protein and proteinuria detected by dipsticks in random urine samples (r = 0.471, p < 0.001). There was a significant positive correlation between proteinuria detected by dipsticks in random urine samples and PCR in both random (r = 0.36, p < 0.001) and 4-hour (r = 0.393, p < 0.001) urine samples. Regression analysis showed that the PCR in random urine samples is a better predictor for significant proteinuria (defined by a total 24-hour protein ≥ 300 mg per day) than both PCR in 4-hour urine samples and proteinuria by dipsticks (adjusted R2 = 0.186, 0.181 and 0.082, respectively). ROC curves were constructed for total 24-hour urine protein, PCR in random and 4-hour urine samples, as predictors of significant proteinuria (≥ 300 mg per day). The area under curve for the ROC curve of PCR in random urine sample was the largest [0.884, 95% CI (0.810 – 0.951)], compared to those of PCR in 4-hour urine sample [0.857, 95% CI (0.783 – 0.931)] and proteinuria by dipsticks [0.665, 95% CI (0.558 – 0.773)], denoting that PCR in random urine samples is a better predictor of significant proteinuria. The best cutoff value for PCR in random urine sample was 0.18 (sensitivity of 90.9% – specificity of 85.3%). The best cutoff value for PCR in 4-hour urine sample was 0.3 (sensitivity of 89.4% – specificity of 61.8%). There was no satisfactory cutoff value for proteinuria detected by dipsticks with satisfactory sensitivity and specificity.
In conclusion, PCR in both random and 4-hour urine correlate well with 24-hour urine protein, but was not justified as a substitute, due to its relatively low specificity.
Recommendations:
We recommend that:
Random PCR and 4 hour PCR coulded be used as a rapid, easy and reliable test for diagnosis of significant proteinuria in hypertensive disorders with pregnancy. But still not justified as a substitute for 24hour urine collection.
Because random PCR and 4 hour PCR are rapid methods for detection of significant proteinuria in preeclampsia they coulded be used in urgent cases.