Hypertensive disorders of pregnancy constitute one of the leading causes of maternal and perinatal mortality and morbidity worldwide. Preeclampsia is estimated to affect 2–8% of pregnancies worldwide [1].

Preeclampsia is a complex multisystemic disease of pregnancy characterized by new-onset hypertension, typically developing after 20 weeks of gestation, often accompanied by new-onset proteinuria or maternal organ dysfunction [2, 3]. Preeclampsia increases the risk of premature birth, perinatal death, and neurodevelopmental delay in the fetus, along with long-term cardiovascular and metabolic diseases. Maternal risks include stroke, cardiovascular disease, and diabetes, imposing a substantial burden on both maternal and fetal health [4, 5, 6, 7].

Despite extensive research, the etiopathogenesis of preeclampsia remains incompletely understood. Therefore, there is no known effective treatment for preeclampsia, and delivery remains the only definitive management [8]. Although proteinuria has been removed as a mandatory criterion for diagnosing preeclampsia, it remains an important diagnostic marker in clinical practice. While 24-hour urine proteinuria (300 mg/day or more) remains the gold standard, some obstetricians may also utilize the protein/creatinine ratio ($>$0.3) in spot urine [9, 10]. Due to fluctuations in urinary protein excretion throughout the day, 24-hour urine collection aims to minimize underdetection of proteinuria. However, this method presents challenges including patient compliance difficulties and diagnostic delays [11].

While previous studies have explored the utility of shorter urine collections, a direct comparison of three consecutive 8-hour intervals within a 24-hour period is less common. This study aimed to not only correlate 8-hour proteinuria with the 24-hour standard but also to identify the optimal 8-hour time window for diagnosis and establish specific, time-dependent Receiver Operating Characteristic (ROC)-derived cutoffs. By doing so, we sought to provide a practical, evidence-based protocol to accelerate the diagnosis of preeclampsia, thereby reducing diagnostic delays and improving patient compliance.

This prospective observational study was conducted at the Department of Obstetrics and Gynecology, Istanbul Training and Research Hospital, Istanbul, Turkey between December 2022 and April 2023. The study protocol was approved by the Ethics Committee of Istanbul Training and Research Hospital (Approval No: 393, Date: 23/12/2022). For analytical purposes, patients were categorized into two groups based on the gold-standard 24-hour urine protein measurement. Patients with $\ge$300 mg/day of proteinuria were classified into the “preeclampsia” group for ROC analysis, as proteinuria is the specific diagnostic marker being evaluated. Patients with 300 mg/day were classified into the “gestational hypertension” (non-preeclampsia) group. The final clinical diagnosis of preeclampsia was made according to American College of Obstetricians and Gynecologists (ACOG) 2022 criteria, which includes hypertension plus either significant proteinuria ($\ge$300 mg/24 h) or, in its absence, new-onset end-organ dysfunction (e.g., thrombocytopenia, renal insufficiency, or elevated liver transaminases). Women with new-onset hypertension and end-organ dysfunction but without proteinuria (300 mg/24 h) were not included in this analysis, as the primary objective was to validate a test for proteinuria itself. Written informed consent was obtained from all participants.

All participants received detailed information about the study and provided written informed consent using an ‘Informed Voluntary Consent Form’ after verbal explanation and adequate time to review the document.

Urine samples were collected in three separate containers over 24 hours in consecutive 8-hour periods, and protein levels were measured in the Istanbul Training and Research Hospital biochemistry laboratory. During the study period, patients’ demographic and clinical data were obtained from direct interviews, medical records, and the clinical database.

Medical history, medication history, obstetric history, and background information were obtained and recorded for each patient with new-onset hypertension. Noninvasive arterial pressure, oxygen saturation (SpO2), and body temperature were measured, and respiratory rates were recorded. After obstetric examination, urine was collected in 3 separate containers for 24 hours in consecutive 8-hour periods (8 AM to 4 PM, 4 PM to 12 AM, and 12 AM to 8 AM) to determine proteinuria levels. Spot urine, complete blood count, urea, creatinine, aminotransferase enzymes, uric acid, lactate dehydrogenase enzyme, and standardized prothrombin time were measured and recorded. Patients were started on an appropriate diet; their blood pressure was monitored four times daily, and values were recorded. Patients were informed about the symptoms of preeclampsia and followed up closely. Urine collection containers were sent to the biochemistry laboratory of Istanbul Training and Research Hospital for proteinuria measurement. Urinary protein concentration was measured using a standardized pyrogallol red–molybdate colorimetric assay on an automated analyzer (Cobas 8000; Roche Diagnostics, Mannheim, Germany). Patients were evaluated according to the diagnostic criteria for preeclampsia. Those who met the criteria were diagnosed with preeclampsia, while others were classified as gestational hypertension (non-preeclampsia). The diagnosis of preeclampsia was based on 24-hour urine proteinuria, which is the gold standard test for detecting urinary proteinuria. The 24-hour urine protein levels were compared with the protein levels in 8-hour urine collections to evaluate the clinical utility of 8-hour urine proteinuria for diagnosing preeclampsia. Preeclampsia was definitively diagnosed per ACOG 2022 criteria: hypertension ($\ge$140/90 mmHg) plus one or more of the following: 24-hour proteinuria $\ge$300 mg/day (gold standard), thrombocytopenia (100,000/µL), renal insufficiency (Cr $>$1.1 mg/dL), elevated transaminases ($\ge$2$\times$Upper Limit of Normal [ULN]), or pulmonary edema. Patients without these criteria were classified as gestational hypertension.

Our sequential 8-hour collection design (rather than randomized groups) was intentionally chosen to: characterize diurnal protein excretion patterns in preeclampsia, enable direct within-patient comparison of shortened vs gold-standard collections, and replicate real-world clinical scenarios where both tests might be run concurrently during diagnostic uncertainty. All participants provided both 8-hour segmented and complete 24-hour samples to control for inter-patient variability.

Three different containers with patient barcodes were prepared to collect urine samples at three different time intervals (8 AM to 4 PM, 4 PM to 12 AM, 12 AM to 8 AM). The urine collection method was explained to the patients by assistant health personnel, and compliance was monitored. Patients were also given a written checklist to record voiding times, and any discrepancies (e.g., missed collections) were documented and excluded from analysis. After the first voided sample was discarded, urine collection started at 8 AM and continued in the same container (bottle number 1) until 4 PM. Then, urine was collected in container number 2 between 4 PM and 12 AM. Finally, urine was collected in container number 3 between 12 AM and 8 AM, completing the collection process. The urine collection process lasted 24 hours in total. After completion, the urine containers were sent to the biochemistry laboratory.

The volumes of the three containers were measured and recorded separately, and 10 cc urine samples were collected from each container. The samples were named Sample-1, Sample-2, and Sample-3 for 8 AM to 4 PM, 4 PM to 12 AM, and 12 AM to 8 AM, respectively. The urine collected over 24 hours was then transferred to a single container and mixed. A 10 cc urine sample was taken from the pooled 24-hour urine sample and named Sample 4. For all four samples, proteinuria levels were calculated in milligrams (Fig. 1).

Fig. 1.

Schematic representation of the sequential 8-hour urine collection method. Urine was collected in three separate, labeled containers corresponding to distinct 8-hour intervals (Bottle 1: 8 AM to 4 PM, Bottle 2: 4 PM to 12 AM, Bottle 3: 12 AM to 8 AM). A fourth sample (Sample 4) was created by pooling aliquots from the three containers to represent the total 24-hour collection.

Patients with documented incomplete urine collections were excluded from the final analysis. All collected data were checked for outliers. Although some patients in the preeclampsia group exhibited very high proteinuria (e.g., $>$7000 mg/day), these values were considered clinically plausible for severe disease and were retained to reflect the real-world spectrum of the condition. A sensitivity analysis excluding values $>$3 standard deviations from the mean did not significantly alter the correlation coefficients or area under the curve (AUC) estimates. All data were analyzed using IBM SPSS (Statistical Package for Social Sciences) version for Windows, version 22.0 (IBM Corp., Armonk, NY, USA). Data analysis began with assessment of statistical assumptions to determine the appropriate tests (parametric/nonparametric). Normality of distribution was assessed using the Kolmogorov-Smirnov test, kurtosis, and skewness values, along with histogram visualization. As the proteinuria data were not normally distributed, as confirmed by the Kolmogorov-Smirnov test, non-parametric tests, including the Spearman correlation coefficient, were used for correlation analyses. ROC analysis was used to assess the diagnostic value of the parameters. The Spearman correlation coefficient was used to examine the relationship between numerical variables. Statistical significance was set at p 0.05. A post-hoc power analysis confirmed that our sample size (n = 99) provided $>$80% power to detect a clinically significant AUC of 0.85 or higher against a null hypothesis of AUC = 0.5, with a two-sided alpha level of 0.05. However, we acknowledge that the sample size may be insufficient for detailed subgroup analyses.

Proteinuria was previously a necessary criterion for the diagnosis of preeclampsia. In 2013, ACOG removed proteinuria as an essential criterion for diagnosing preeclampsia. However, in the absence of severe end-organ damage, urinary protein quantification remains an important criterion for evaluating hypertension. The gold standard test for measuring protein excretion in pregnancy is 24-hour urine protein measurement. However, this method is not practical, especially in cases requiring rapid diagnosis. Additionally, 24-hour urine collection may show limitations due to factors such as inadequate collection, inconvenience, and spillage during collection.

While randomization between 8-hour and 24-hour collection groups could reduce carryover effects, our paired design offered critical advantages: First, it demonstrated that 8-hour samples can reliably predict 24-hour results in the same patient—the fundamental clinical question when considering test substitution. Second, it revealed time-dependent performance variations (daytime vs nighttime) that would be obscured in parallel-group designs. Third, this approach matches clinical practice where abbreviated collections often precede 24-hour confirmation. We acknowledge that future randomized implementation studies could further validate our findings.

A study by Adelberg et al. [12] in 2001 involved 65 pregnant women with hypertensive disorders. For each participant, urine was collected over a 24-hour period, segmented into distinct containers for the initial 8 hours, the subsequent 4 hours, and the final 12 hours. The researchers then quantified urine volume and total protein concentrations in the 8-hour, 12-hour, and full 24-hour samples. Employing simple regression analysis, they assessed the relationship between the 8-hour and 12-hour measurements against the established 24-hour findings. This analysis revealed substantial correlations between both the 8-hour and 12-hour urine collections and the gold-standard 24-hour samples. Furthermore, even in patients without proteinuria, a significant correlation persisted between the 12-hour and 24-hour urine collections [12].

In the study conducted by Wongkitisophon et al. [13] in 2003, 38 patients diagnosed with hypertensive disease of pregnancy were included. In this study, urine samples were collected in 2 separate periods over 28 hours: the first 4 hours and the subsequent 24 hours. In these patients with hypertensive disorders of pregnancy, a correlation was found between the results of the 4-hour urine protein and the results of the 24-hour urine protein (p 0.001) [13].

Schubert and Abernathy demonstrated a significant correlation between proteinuria in 12- and 24-hour urine samples for the diagnosis of preeclampsia. They concluded that a protein/creatinine ratio of 0.15 ruled out significant proteinuria [14].

In the study conducted by Silva et al. [15] in 2018, 99 pregnant patients were included. The diagnosis of preeclampsia was confirmed in 42 of these patients (42%). Urine samples were collected from the patients in 2 separate containers for 24 hours in consecutive 12-hour periods. Proteinuria cut-off values for the diagnosis of preeclampsia were 150 mg/12 hours for 12-hour urine and 300 mg/day for 24-hour urine. Qualitative analysis (preeclampsia or no preeclampsia) showed significant agreement between the 12- and 24-hour samples (Cohen $\kappa$: 0.779). Sensitivity was 85.9% (95% confidence interval [CI]: 81–90%), and specificity was 91.7% (95% CI: 88%–95%). There was no statistically significant difference between the two 12-hour urine samples [15].

In 2014, Rani et al. [16] conducted research involving 125 pregnant patients who were diagnosed with preeclampsia via a dipstick test indicating urine albumin levels greater than +1. Their methodology involved collecting urine samples over a 24-hour period, segmented into five different time intervals. The total protein from the two-hour, four-hour, eight-hour, and 12-hour collections was quantified and subsequently compared with the total 24-hour protein measurement. Using Pearson’s correlation coefficient, the analysis revealed a significant correlation (p 0.01) between the 24-hour total and the shorter collection periods, with high correlation coefficients of 0.97 (2-hour), 0.97 (4-hour), 0.96 (8-hour), and 0.97 (12-hour), respectively. The study also established diagnostic cut-off values of 25 mg, 50 mg, 100 mg, and 150 mg for the 2, 4, 8, and 12-hour samples. At these thresholds, the corresponding sensitivity rates were 92.4%, 95.2%, 91.5%, and 96.2%, while the specificity rates were 68.4%, 94.7%, 84.2%, and 84.2%, respectively [16]. Our investigation revealed a strong positive correlation between 8-hour and 24-hour urine protein measurements. Furthermore, no statistically significant disparities were observed among the different 8-hour urine collection periods, for which distinct cut-off values were also determined.

Kieler et al. [17] compared urine albumin in spot and 12-hour urine samples with 24-hour urine collection in 30 women with preeclampsia. They found that the 12-hour collection showed a good correlation with the 24-hour collection, but the relationship between spot urine and protein content in the 24-hour urine was poor. They concluded that 12-hour urine collection can be used instead of 24-hour urine collection, which is the gold standard for detecting proteinuria, while spot urine gives erroneous results [17].

A study by Haghighi et al. [18] supports the use of shorter urine collection times as an alternative to the 24-hour standard. They found a significant positive correlation between 4-hour, 8-hour, and 12-hour (daytime/nighttime) samples and the 24-hour results. This led them to conclude that these shorter tests are effective for detecting proteinuria in pregnant women with new-onset hypertension when a rapid diagnosis is needed [18]. Our findings indicate that an 8-hour urine collection is a dependable substitute for the 24-hour test in detecting proteinuria among pregnant patients with new-onset hypertension. The 8 AM to 4 PM collection period, in particular, demonstrated the strongest agreement with the 24-hour results. Our findings align with a growing body of evidence supporting shorter collection periods and contribute to the ongoing debate about the optimal rapid diagnostic method. For instance, a recent meta-analysis by Tian et al. [19] found that 12-hour proteinuria estimation showed slightly better diagnostic performance (AUC: 0.97) than the spot urine protein-to-creatinine ratio (uPCR) (AUC: 0.93) when compared to the 24-hour standard. More directly comparable to our study, Massalha et al. [20] investigated four different short methods and also confirmed strong correlations with the 24-hour method. Our study builds upon this work by specifically characterizing three distinct 8-hour intervals, identifying the 8 AM to 4 PM window as the most accurate (AUC: 0.95), performing on par with the 12-hour collections reported by Tian et al. [19] and offering a practical daytime alternative for clinical triage. It is important to note that while the ROC curve for the 24-hour collection is presented for illustrative comparison, its AUC value should be interpreted with caution. Since 24-hour proteinuria was used as the gold standard to define the diagnostic outcome, its performance analysis is inherently circular. The primary utility of our ROC analysis lies in evaluating the performance of the shorter 8-hour collection intervals against this established benchmark.

In our analysis, we proposed a standardized 100 mg/8 hours cutoff for clinical utility, although ROC analysis identified slightly different optimal thresholds for each time window (ranging from 71 to 121 mg). Our rationale for adopting a single, pragmatic threshold was threefold. First, 100 mg directly corresponds to one-third of the established 24-hour diagnostic criterion of 300 mg, making it conceptually intuitive for clinicians [21]. Second, this value maintained acceptable diagnostic performance across all time intervals, particularly during daytime hours, which is when most clinical decisions are made. Third, implementing a single, memorable cutoff simplifies clinical protocols and reduces the potential for error compared to using multiple time-dependent thresholds. While a lower cutoff for nighttime collections could slightly increase sensitivity, we believe the substantial benefit of clinical simplicity justifies the use of a unified 100 mg threshold for initial screening [22].

It is also important to position our findings within the context of other rapid diagnostic alternatives, most notably the spot urine protein-to-creatinine ratio (uPCR). While uPCR offers the advantage of near-instantaneous results from a single void, its accuracy can be affected by variations in urine concentration and diurnal creatinine excretion [23]. The 8-hours collection, though requiring more time than a spot test, offers a more robust measure of protein excretion over a defined period, potentially mitigating the variability seen with single spot samples. Our proposed method could serve as a valuable intermediate step: faster than a 24-hour collection but potentially more reliable than a single uPCR, especially in cases where the initial spot result is equivocal.

In a clinical setting, the 8-hour urine collection protocol could be seamlessly integrated into obstetric triage units. Upon presentation with new-onset hypertension, a patient could immediately begin an 8-hour collection, most practically during daytime working hours (8 AM to 4 PM). A result exceeding our ROC-derived cutoff of 121 mg could prompt expedited clinical decision-making, such as hospital admission or initiation of treatment, without waiting a full 24 hours. This method would not necessarily replace the 24-hour collection but could serve as a rapid triage tool to identify high-risk patients much earlier, reserving the full 24-hour test for cases where the 8-hour result is borderline or for longitudinal monitoring.

Our findings also raise an important question regarding clinical implementation: whether specific decision thresholds for preeclampsia should be explicitly noted on laboratory reports. Currently, most laboratories report a general upper reference limit for proteinuria (e.g., 150 mg/24 h), which is not specific to pregnancy [24, 25]. As argued by Hortin, incorporating pregnancy-specific decision limits, such as a 121 mg/8 hours threshold, could significantly improve diagnostic clarity for obstetricians and reduce the risk of misinterpretation [26]. However, implementing such changes would require greater harmonization of urine protein assays to ensure that numerical cutoffs are transferable between different clinical sites [27].

Our study demonstrates that 8-hour urine collection is a reliable alternative to 24-hour urine collection for assessing proteinuria in pregnant women with new-onset hypertension. Specifically, proteinuria measured in urine collected between 8 AM and 4 PM showed the strongest alignment with the 24-hour gold standard for diagnosing preeclampsia. Furthermore, a statistically significant strong positive correlation was found between all 8-hour urine collections and the 24-hour measurements. Overall, this study suggests that 8-hour urine collection can serve as a valuable alternative method to assess proteinuria, thereby minimizing diagnostic delays and patient non-compliance often associated with 24-hour urine collection, which is crucial for timely diagnosis and management of preeclampsia. However, while our single-center sample size (n = 99) provided adequate power for correlation analyses (post-hoc power $>$80%), these findings require validation in larger, multi-center cohorts across diverse populations to confirm the generalizability of 8-hour proteinuria cut-off values, particularly for preeclampsia subtypes (e.g., early-onset vs. late-onset). Future studies should aim for $\ge$500 participants with balanced representation of ethnic/geographic groups.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

IIO: Development of study concept and design, data collection and analysis, manuscript writing, and critical revision. ESG: Study design, interpretation of results, critical revision, and approval of the manuscript. YZK: Statistical analysis of data, scientific content review. Contribution to manuscript drafting and critical revision. ESK, AGZ: Data interpretation, contribution to manuscript writing, and final review. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study protocol was approved by the Ethics Committee of Istanbul Training and Research Hospital (Approval No: 393, Date: 23/12/2022). The study was conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participants prior to their inclusion in the study.

We would like to express our gratitude to all the patients who participated in this study and to the clinical and laboratory staff of the Istanbul Training and Research Hospital for their invaluable support. We also thank the peer reviewers for their insightful comments and suggestions that helped improve this manuscript.

This research received no external funding.

The authors declare no conflict of interest.