Risk Stratification and Safety Profile of Pharmacological Prophylaxis for Venous Thromboembolism in Postpartum Women Following Cesarean Section: A Comparative Evaluation

Abstract

Objective

Venous thromboembolism (VTE) remains a leading cause of maternal morbidity and mortality worldwide. This study evaluated VTE risk stratification among postpartum women following cesarean section, compared the 2023 National Guidelines (NG) with the Caprini score, and assessed pharmacological thromboprophylaxis safety.

Methods

A single-center prospective cohort study was conducted on 386 women undergoing cesarean delivery at Nhan Dan Gia Dinh Hospital between February and July 2025. VTE risk was independently assessed using the 2023 NG and Caprini score. Demographic characteristics, operative variables, and short-term inpatient outcomes related to low-molecular-weight heparin (LMWH) or unfractionated heparin prophylaxis were analyzed.

Results

According to the 2023 NG, 71.0% of patients were classified as low risk, 15.5% as intermediate risk, and 13.5% as high risk. Major risk factors included emergency cesarean delivery (45.3%), maternal age above 35 years (23.1%), and obesity (20.5%). Using the Caprini score, 45.3% met criteria for pharmacological prophylaxis, representing a 56.2% increase compared with the 2023 NG. Among 110 women receiving thromboprophylaxis, LMWH accounted for 99.1% of treatment. No severe hemorrhagic complications or thrombocytopenia were observed. Adverse events were mild and localized, mainly injection-site ecchymosis.

Conclusion

The 2023 NG generated fewer pharmacological prophylaxis indications than the Caprini score in this cohort. LMWH prophylaxis showed favorable short-term inpatient safety after cesarean delivery. Because follow-up was limited to hospitalization and the study was not powered for rare VTE events, these findings should be interpreted as allocation patterns and short-term safety rather than comparative effectiveness.

Key Points

• ● The 2023 National Guidelines identified 29.0% of post-cesarean women at moderate-to-high VTE risk, resulting in fewer pharmacological prophylaxis indications than the Caprini score during inpatient postpartum care.

• ● The Caprini score generated a higher proportion of pharmacological prophylaxis indications than the 2023 NG in this post-cesarean cohort.

• ● Subcutaneous low-molecular-weight heparin (LMWH) prophylaxis demonstrated favorable short-term inpatient tolerability in this postpartum cohort.

• ● No severe hemorrhagic events or thrombocytopenia were recorded among patients receiving postpartum pharmacological prophylaxis.

• ● Specialized obstetric risk models combined with active nursing surveillance may improve consistency of risk recognition and postpartum care pathways.

Introduction

Venous thromboembolism (VTE), comprising deep vein thrombosis (DVT) and pulmonary embolism (PE), is recognized as a leading cause of severe morbidity and mortality in women of reproductive age worldwide. ¹ Women during pregnancy and the postpartum period have a significantly higher risk of developing VTE compared to non-pregnant women of similar demographics. Epidemiological data indicate that the incidence of pregnancy-related VTE ranges around 1.5 cases per 1,000 deliveries, with PE accounting for nearly one-third of these events and acting as the direct cause of approximately 10% of maternal deaths in developed nations.^1,2 This risk increases progressively throughout pregnancy, reaches its peak during delivery, and remains elevated during the first 6 weeks of the puerperium. ³

The prevalence of contemporary risk factors is altering the epidemiological profile of VTE. Notably, there is a global surge in the rate of cesarean sections (CS). CS has been identified as an independent risk factor, increasing the risk of postpartum VTE by 2 to 4 times compared to vaginal delivery.^2,4 The combination of extensive tissue trauma, pelvic vascular injury, and prolonged postoperative immobilization establishes an environment conducive to thrombus formation, sometimes necessitating advanced mechanical prophylaxis like inferior vena cava filters.^5,6 Sedentary postoperative behavior and prolonged immobilization further elevate the cumulative VTE risk in obstetric cohorts. ⁷

To address this condition, several medical societies worldwide have developed Risk Assessment Models (RAMs) to stratify risk and guide prophylactic interventions. The most prominent include guidelines from the Royal College of Obstetricians and Gynaecologists (RCOG), the American College of Chest Physicians (ACCP), and the American College of Obstetricians and Gynecologists (ACOG).^8,9 However, significant fragmentation exists among these clinical recommendations regarding intervention thresholds. For instance, ACCP applies a stricter threshold, indicating prophylaxis for only about 8% of patients.^8,10 In Vietnam, Decision No. 3908/QD-BYT, issued in 2023, established a comprehensive National Guideline (NG) for VTE prophylaxis, which includes a dedicated obstetrics module adapted to suit the specific physical and epidemiological characteristics of Asian women. ¹¹

Before the widespread implementation of the 2023 NG, many healthcare facilities utilized the Caprini score for risk stratification. The Caprini score is a well-established and highly validated tool in general, orthopedic, and neurological surgery. ¹¹ Nevertheless, it was not specifically developed for obstetric patients, and its weighting of age, surgical exposure, and comorbidities may not fully capture pregnancy-specific mechanisms or Asian postpartum epidemiology. Clinical implementation of such non-specific risk assessment tools in postpartum patients often leads to unsatisfactory prevention or clinical mismatch. ¹² Furthermore, physical and mobility limitations in postpartum patients represent a significant independent risk. ¹³ Evaluating a surgical scoring system within an obstetric population may lead to risk classification patterns that differ from obstetric-specific assessment models ¹⁴ , potentially influencing prophylaxis recommendations. ⁹

Pregnancy and the early puerperium create a prothrombotic state through physiological hypercoagulability, venous stasis, and endothelial injury. Cesarean delivery further amplifies this risk through surgical tissue trauma, reduced mobility, and perioperative inflammation. These mechanisms justify structured postpartum VTE risk assessment but do not replace clinically calibrated risk models.^3,15

This study aimed to determine the prevalence of VTE risk categories among postpartum women following CS and to compare prophylaxis allocation using the 2023 NG and the Caprini score. The Caprini score was used as a pragmatic comparator because it has been widely adopted in surgical populations; however, it was not developed for obstetric patients and should not be considered a reference standard for postpartum VTE prediction. The study also evaluated the feasibility and short-term inpatient safety of pharmacological prophylaxis with LMWH. The findings are intended to inform context-specific postpartum thromboprophylaxis pathways in resource-limited clinical settings.

Methods

Study Design and Population

This study was designed as a single-center prospective cohort study, conducted at the Obstetrics Department of Nhan Dan Gia Dinh Hospital. Data collection spanned from February 2025 to July 2025. The hospital serves as a tertiary referral center receiving a high volume of complex pregnancies and obstetric emergencies.

The target population included postpartum women aged 18 years and older who were indicated for a cesarean section. Exclusion criteria comprised patients with psychiatric disorders incapable of providing informed consent, patients transferred before completing the monitoring protocol, and those who declined participation. During the study period, there were 870 total deliveries at the hospital, of which 395 were cesarean sections. After excluding 2 patients who declined participation and 7 patients under 18 years of age, 386 valid medical records were successfully collected. This represents an inclusion rate of 97.7% of all eligible cesarean cases, establishing a high degree of sampling reliability. Clinical outcomes were monitored only during the index postpartum hospitalization, which typically lasted 3 to 5 days after cesarean delivery. The study did not include systematic outpatient surveillance after discharge and therefore was not designed to detect VTE or bleeding events occurring during the full 6-week postpartum period.

The participant selection process is summarized in Figure 1.

Figure 1.

Study flowchart. Flow diagram illustrating participant selection, study inclusion, risk assessment according to the 2023 Vietnamese National Guidelines and the Caprini score, pharmacological thromboprophylaxis allocation, and inpatient follow-up

Sample Size Determination and Data Collection

The minimum sample size was determined based on the formula for estimating a proportion in a finite population. Because no reliable local estimate of prophylaxis indication rates was available, p = 0.5 was used to maximize variance and ensure an adequate sample size. The calculated sample size was 385 cases, as derived from the classic formula:

$$n={\mathrm{{\rm Z}}}^2\frac{p\left(1-p\right)}{d^2}$$

The continuous convenience sampling process successfully collected 386 valid medical records. Objective clinical information was extracted from electronic medical records, hematological laboratory results, and surgical reports. Simultaneously, researchers conducted direct interviews with the patients using a standardized questionnaire 24-48 hours postoperatively. The coded study variables included demographics, pre-pregnancy Body Mass Index (BMI) categorized according to World Health Organization (WHO) standards for Asians, medical history, and obstetric/perioperative parameters. ¹¹ Additional variables available from the thesis database and incorporated into the revised analysis included parity, previous cesarean delivery, preterm birth, neonatal birth weight, estimated blood loss, postpartum hemorrhage, preeclampsia, assisted reproduction, multiple pregnancy, stillbirth, documented medical comorbidities, and recorded thrombophilia-related risk items.

Risk Stratification Assessment and Intervention Protocol

Each patient underwent a parallel and independent assessment using two risk stratification systems.

• • 2023 National Guidelines (NG): Based on Decision No. 3908/QD-BYT, this system categorizes factors into pre-existing risks, obstetric risks, and transient risks. The total score is classified into: Low risk (≤2 points), Intermediate risk (3 points), and High risk (≥4 points). ¹¹

• • Caprini Score: Allocates points across various clinical factors. Risk thresholds are divided into: Low (0-2 points), Moderate (3-4 points), High (5-8 points), and Highest (>8 points). ¹⁶

The structure and scoring rules of the 2023 Vietnamese National Guideline VTE risk assessment model are summarized in Table 1.

Table 1.

Summary of the 2023 Vietnamese National Guideline VTE Risk Assessment Model

Risk domain	Risk factor	Score
Pre-existing	Previous VTE	4
Pre-existing	Previous VTE provoked by major surgery; high-risk thrombophilia; major comorbidity including malignancy, heart failure, active systemic lupus erythematosus, active pulmonary tuberculosis, nephrotic syndrome, diabetes with renal complications, or moderate/severe thalassemia after splenectomy	3
Pre-existing	Obesity, BMI ≥40 kg/m2	2
Pre-existing	Obesity, BMI 30–39 kg/m2; smoking ≥10 cigarettes/day; family history of VTE; low-risk thrombophilia; maternal age >35 years; parity ≥3; large varicose veins	1
Obstetric	Cesarean delivery during labor	2
Obstetric	Elective cesarean delivery; preeclampsia; assisted reproduction/IVF; multiple pregnancy; instrumental delivery; prolonged labor >24 hours; postpartum hemorrhage ≥1000 mL or requiring transfusion; stillbirth; preterm birth <37 weeks	1
Transient	Ovarian hyperstimulation syndrome in the first trimester	4
Transient	Any surgery or procedure during pregnancy or puerperium except episiotomy repair, appendectomy, or sterilization; severe vomiting	3
Transient	Immobility/dehydration; systemic infection or postpartum infection	1
Risk categories	Low risk ≤2 points; intermediate risk 3 points; high risk ≥4 points	​

Patients classified into the Intermediate and High-risk groups according to the 2023 NG received pharmacological prophylactic intervention. The standard protocol utilized LMWH (Enoxaparin) 40 mg via subcutaneous injection (SC) once daily. For specific comorbidities, adjustments utilizing Unfractionated Heparin (UFH) were made. The timing of treatment initiation strictly adhered to anesthetic safety principles. ¹¹ Severe hemorrhage was defined as bleeding causing hemodynamic instability, requiring blood transfusion, reoperation, intensive care admission, or discontinuation of anticoagulant prophylaxis. Clinically relevant non-major bleeding was defined as overt bleeding not meeting criteria for severe hemorrhage but requiring medical evaluation, additional laboratory testing, local intervention, or temporary withholding of prophylaxis. Minor bleeding was defined as self-limited bleeding or localized injection-site ecchymosis not requiring medical intervention.

Statistical Analysis

Data were analyzed using Stata 14 software. Quantitative variables were expressed as means and standard deviations. Categorical variables were summarized using frequencies, percentages, and 95% Confidence Intervals (95% CI). The Chi-square test was applied to evaluate the statistical significance of differences between risk stratifications and clinical characteristics. The level of statistical significance was set at p<0.05.

Results

Baseline Characteristics and Anthropometric Factors

The study sample comprised 386 postpartum women with a mean age of 31.4 ± 5.7 years. Specifically, 89 women (23.1%) were in the > 35 age group. Geographically, 49.7% resided in Ho Chi Minh City, with the remainder from suburban provinces. Most participants were of Kinh ethnicity (98.2%). Primiparous women accounted for 162 cases (41.9%), and 67 women (17.4%) had at least two previous births. First cesarean delivery accounted for 202 cases (52.3%). Normal previous obstetric history was recorded in 151 women (39.1%), while previous miscarriage, induced abortion, and stillbirth were recorded in 76 (19.7%), 89 (23.1%), and 60 (15.5%) women, respectively. Medical comorbidities were recorded in 70 women (18.1%), including cardiovascular disease in 27 (7.0%), respiratory disease in 8 (2.1%), thyroid disease in 9 (2.3%), diabetes or renal disease in 11 (2.8%), and other conditions in 15 (3.9%). Nutritional status and pre-pregnancy BMI demonstrated a significant prevalence of overweight and obesity.

The distribution of pre-pregnancy BMI categories according to Asian WHO criteria is presented in Table 2.

Table 2.

Distribution of Pre-pregnancy BMI According to WHO Criteria for the Asian Region (Overall Distribution p<0.001*)

BMI classification	Total (n=386) (%)	95% CI	Low risk (n=274)	Prophylaxis indicated (n=112)
Underweight	16 (4.1)	2.4 - 6.6	16	0
Normal	196 (50.8)	45.7 - 55.8	180	16
Pre-obese	95 (24.6)	20.4 - 29.2	60	35
Obesity Class I	57 (14.8)	11.4 - 18.7	18	39
Obesity Class II	20 (5.2)	3.2 - 7.9	0	20
Obesity Class III	2 (0.5)	0.1 - 1.9	0	2

*p-value < 0.001 calculated using Chi-square test for the overall distribution of BMI Classification between the Low Risk and Prophylaxis Indicated groups.

The data indicate that 20.5% of the women were categorized as obese (BMI ≥25 kg/m²), of which 5.7% had Class II and III obesity. Associated medical conditions were noted in 18.1% of patients, with cardiovascular diseases accounting for 7.0%.

Available group-stratified baseline and clinical characteristics according to the 2023 National Guidelines are shown in Table 3.

Table 3.

Baseline and Clinical Characteristics With Available Group-Stratified Data According to the 2023 National Guidelines

Characteristic	Category or measure	Total n (%)	Low risk n (%)	Prophylaxis indicated n (%)	p-value
BMI category	Underweight/normal/pre-obese/obese I/obese II/obese III	16 (4.1)/196 (50.8)/95 (24.6)/57 (14.8)/20 (5.2)/2 (0.5)	16/180/60/18/0/0	0/16/35/39/20/2	<0.001
Obesity	BMI ≥25 kg/m2	79 (20.5)	18 (6.6)	61 (54.5)	<0.001
Preterm birth	<37 weeks	37 (9.6)	0 (0.0)	37 (33.0)	<0.001
Emergency CS	Yes	175 (45.3)	104 (38.0)	71 (63.4)	<0.001

Abbreviations: CS, cesarean section; NG, National Guidelines. Data are presented as n (%) unless otherwise indicated. P-values were calculated using the Chi-square test. Variables are shown where valid group-stratified data were available.

Obstetric History and Surgical Characteristics

Regarding reproductive history, 41.9% were primiparous, and 17.4% belonged to the multiparous group (≥2 previous births). In terms of surgical history, 52.3% of the women underwent CS for the first time. The nature of the cesarean section represents a critical variable in triggering the endogenous coagulation cascade. Pregnancy outcome characteristics and the nature of cesarean delivery are summarized in Table 4.

Table 4.

Pregnancy Outcome Characteristics and Nature of Surgery

Characteristic	Category	Total n (%)	Low risk n (%)	Prophylaxis indicated n (%)	p-value
Gestational age	Preterm (<37 weeks)	37 (9.6)	0 (0.0)	37 (33.0)	<0.001
	Term (≥37 weeks)	349 (90.4)	274 (100.0)	75 (67.0)	​
Method of surgery	Elective CS	211 (54.7)	170 (62.0)	41 (36.6)	<0.001
	Emergency CS	175 (45.3)	104 (38.0)	71 (63.4)	​

The rate of emergency CS after the onset of labor constituted 45.3%. The mean neonatal birth weight was 3141 ± 509 g, with 7 newborns (1.8%) weighing at least 4000 g. Estimated blood loss was <1000 mL in 383 women (99.2%) and ≥1000 mL in 3 women (0.8%). These 3 cases met the study definition of postpartum hemorrhage because they required blood transfusion. No reoperation, balloon tamponade, intensive care admission, or anticoagulant discontinuation due to hemorrhage was documented in the inpatient safety dataset. The observed PPH rate was lower than rates reported in some contemporary cesarean cohorts, which may reflect local case-mix, routine active management of the third stage of labor, and the fact that our study recorded PPH using a threshold of ≥1000 mL or transfusion rather than broader definitions based on estimated blood loss ≥500 mL. Perioperative bleeding and postpartum hemorrhage characteristics are detailed in Table 5.

Table 5.

Perioperative Bleeding and Postpartum Hemorrhage Profile

Variable	Total cohort n (%) or value	Comment
Estimated blood loss	Median 350 mL	Recorded intraoperatively
Estimated blood loss <1000 mL	383 (99.2)	No PPH threshold reached
Estimated blood loss ≥1000 mL	3 (0.8)	Meets PPH threshold after cesarean delivery
PPH requiring blood transfusion	3 (0.8)	All recorded PPH cases required transfusion
Severe hemorrhage among prophylaxis recipients	0/110 (0.0)	No severe hemorrhage occurred among prophylaxis recipients during inpatient follow-up.
Clinically relevant non-major bleeding among prophylaxis recipients	0/110 (0.0)	No medical intervention or temporary withholding of prophylaxis

Abbreviations: PPH, postpartum hemorrhage. Severe hemorrhage and clinically relevant non-major bleeding were defined as described in the Methods.

The observed PPH rate reflects clinically significant hemorrhage defined as blood loss ≥1000 mL and/or transfusion-requiring bleeding; milder bleeding episodes were not systematically captured. This may partly explain the lower PPH frequency compared with some published cesarean cohorts.

VTE Risk Stratification: 2023 National Guidelines vs. Caprini Score

The distribution of risk factors according to the 2023 NG revealed that the variables with the highest frequencies included elective CS (54.7%), emergency CS (45.3%), maternal age > 35 (23.1%), and multiparity (17.4%). Additional obstetric risk factors included preeclampsia in 32 women (8.3%), preterm birth in 37 women (9.6%), assisted reproduction or in vitro fertilization in 5 women (1.3%), multiple pregnancy in 6 women (1.6%), stillbirth in 1 woman (0.3%), and postpartum hemorrhage in 3 women (0.8%). No participant had documented previous VTE, family history of VTE, or recorded high- or low-risk thrombophilia in routine clinical records.

The comparison of VTE risk stratification between the 2023 NG and the Caprini score is presented in Table 6.

Table 6.

Comparison of VTE Risk Stratification Between the 2023 National Guidelines and the Caprini Score

Risk stratification	2023 NG (n=386) (%)	95% CI	Caprini score (n=386) (%)	95% CI	p-value
Low Risk	274 (71.0)	66.2 - 75.4	211 (54.7)	49.6 - 59.7	< 0.001
Intermediate Risk	60 (15.5)	12.1 - 19.5	166 (43.0)	38.0 - 48.1	​
High Risk	52 (13.5)	10.2 - 17.3	9 (2.3)	1.1 - 4.4	​
Total indicated	112 (29.0)	24.6 - 33.8	175 (45.3)	40.3 - 50.4	< 0.001

The two assessment models generated significantly different prophylaxis allocation patterns. According to the 2023 NG, 112 of 386 women (29.0%) met criteria for pharmacological prophylaxis, compared with 175 of 386 women (45.3%) according to the Caprini score. Cross-tabulation showed that 98 of 274 women (35.7%) classified as low risk by the 2023 NG would have been reclassified as requiring pharmacological prophylaxis under the Caprini score.

Safety Profile of Pharmacological Intervention

Of the 112 patients indicated for treatment under the 2023 NG, 110 people consented to participate in the protocol. The preferred pharmacological agent was LMWH (Enoxaparin), utilized by 99.1% of the treated cohort (109/110 cases). Characteristics of pharmacological prophylaxis and inpatient adverse events among treated women are summarized in Table 7.

Table 7.

Characteristics and Inpatient Adverse Events of Pharmacological Prophylaxis Recipients (n=110)

Evaluation factor	Intermediate risk (n=59) n (%)	High risk (n=51) n (%)	Overall n (%) [95% CI]	p-value
Type of drug used
LMWH (Enoxaparin)	59 (100.0)	50 (98.0)	109/110 (99.1); 95% CI 95.0–99.9	0.279
UFH	0 (0.0)	1 (2.0)	1/110 (0.9); 95% CI 0.0–5.0	​
Bleeding and adverse events
No complications	48 (81.4)	42 (82.4)	90/110 (81.8); 95% CI 73.8–88.2	0.893
Any local adverse reaction	11 (18.6)	9 (17.6)	20/110 (18.2); 95% CI 11.8–26.2	0.893
Minor bleeding: injection-site ecchymosis	5 (8.5)	5 (9.8)	10/110 (9.1); 95% CI 4.6–15.6	​
Localized pain	4 (6.8)	4 (7.8)	8/110 (7.3); 95% CI 3.3–13.3	​
Localized erythema/pruritus	2 (3.3)	0 (0.0)	2/110 (1.8); 95% CI 0.2–6.4	​
Clinically relevant non-major bleeding	0 (0.0)	0 (0.0)	0/110 (0.0); 95% CI 0.0–3.3	​
Severe hemorrhage	0 (0.0)	0 (0.0)	0/110 (0.0); 95% CI 0.0–3.3	​
Thrombocytopenia	0 (0.0)	0 (0.0)	0/110 (0.0); 95% CI 0.0–3.3	​
Inpatient DVT or PE	0 (0.0)	0 (0.0)	0/110 (0.0); 95% CI 0.0–3.3	​

Abbreviations: CI, confidence interval; DVT, deep vein thrombosis; LMWH, low-molecular-weight heparin; PE, pulmonary embolism; UFH, unfractionated heparin. P-values compare intermediate-risk and high-risk groups using the Chi-square test or Fisher’s exact test, as appropriate. For zero-event outcomes, the upper 95% confidence limit was estimated using the exact binomial method.

Pharmacological prophylaxis demonstrated favorable short-term inpatient safety among treated women. Overall, 90 of 110 women (81.8%) experienced no recorded complications, while 20 (18.2%) had local adverse reactions, including injection-site ecchymosis in 10 (9.1%), localized pain in 8 (7.3%), and localized erythema or pruritus in 2 (1.8%). Injection-site ecchymosis was classified as minor bleeding because it was localized and did not require medical intervention. No clinically relevant non-major bleeding, severe hemorrhage, thrombocytopenia, DVT, or PE was detected during inpatient follow-up. Because adverse-event ascertainment ended at discharge, these findings should not be extrapolated to the full postpartum prophylaxis period.

Discussion

This study delineates the epidemiological parameters of VTE risk factors among women undergoing CS in Vietnam, providing real-world implementation data regarding the application of the 2023 National Guidelines in a Vietnamese tertiary obstetric center. The analytical results indicate that optimizing the allocation of medical resources and pharmacological interventions necessitates a specialized tool tailored for the obstetric population. However, the present comparison reflects prophylaxis allocation patterns rather than predictive validity or comparative clinical effectiveness. The absence of VTE events in our sample precludes a direct comparison of efficacy; however, the clinical data support favorable acute inpatient safety of the protocol.

The high inclusion rate reflects active participation by trained nursing and obstetric staff in identifying eligible patients, interviewing participants 24–48 hours after surgery, and completing risk assessment documentation. This workflow is clinically relevant because effective VTE prevention depends not only on the risk model itself but also on reliable implementation during routine postpartum care.

Epidemiological Dynamics and Hematological Risk Factors

The demographic data reflect contemporary obstetric trends in an urban tertiary setting. The older maternal age group (>35 years) accounted for 23.1%, supporting its inclusion as a clinically relevant factor in structured postpartum VTE assessment.

Concurrently, overweight and obesity were prevalent in 20.5% of the women. Obesity is an established independent risk factor for postpartum VTE, and its incorporation using Asian BMI thresholds is clinically relevant in this population. ¹⁷

The nature of surgery was also important. Emergency CS accounted for 45.3% of procedures and may reflect a higher-risk perioperative context than elective CS because of labor exposure, surgical urgency, and reduced opportunity for preoperative optimization. ⁴

Preeclampsia, preterm birth, assisted reproductive technologies, and multiple pregnancy are clinically relevant obstetric factors included in several obstetric-specific risk frameworks.^17-19 In the present cohort, preterm birth occurred in 9.6% of women and was observed only among those with prophylaxis indications.

Optimizing Assessment Tools and Pharmacoeconomic Strategies

The Caprini score is recognized as a highly validated tool in general and orthopedic surgery. ⁷ However, it was not specifically developed for obstetric patients and gives substantial weight to age, surgery, and general comorbidity rather than pregnancy-specific factors. In this cohort, the Caprini score generated a higher proportion of pharmacological prophylaxis indications than the 2023 NG, 45.3% versus 29.0%. This finding should be interpreted as a difference in allocation rather than evidence of differential predictive accuracy, because the study was not powered or followed long enough to evaluate either model against thrombotic outcomes. Therefore, the present study should not be interpreted as a validation study of either risk assessment model.

International guidelines exhibit substantial variance in establishing intervention thresholds. Studies evaluating the ACCP guidelines demonstrate that they identify approximately 8.3% to 16.6% of the postpartum population as requiring pharmacological prophylaxis, adhering to stringent criteria.^6,20 In contrast, the RCOG guidelines encompass a broader sensitivity, classifying between 53.6% and 73.8% of postpartum women as intermediate or high risk, thereby significantly expanding the indications for LMWH.^6,20 Within this spectrum, the 2023 NG generated fewer prophylaxis indications than the Caprini score in our cohort and produced a narrower prophylaxis allocation pattern in this setting. The 2023 NG accounts for contemporary factors such as IVF and uses obesity thresholds adapted to Asian populations, while its impact on VTE reduction requires evaluation in adequately powered studies with complete postpartum follow-up.

The implementation of a multidisciplinary nurse-assisted early warning intervention has been shown to be feasible in clinical practice. In a large-scale cohort of over 27,000 deliveries, the active involvement of trained nursing staff in real-time electronic risk screening was associated with a reduction in symptomatic postpartum VTE from 2.4 per 1,000 deliveries in the control group to 0.29 per 1,000 in the intervention group (p<0.001). This evidence supports the integration of specialized RAMs into electronic medical record (EMR) systems, backed by active nursing surveillance as a dual-layer safety net. ²¹

Mechanical Prophylaxis and Postpartum Care Pathway

Mechanical prophylaxis, including early mobilization, graduated compression stockings, and intermittent pneumatic compression, remains an important component of postpartum VTE prevention, particularly for women with contraindications to anticoagulation or transiently increased bleeding risk^.2,3 In routine care at our center, early ambulation and avoidance of dehydration are encouraged after cesarean delivery, while device-based mechanical prophylaxis is applied selectively according to clinical judgment and availability. Device-based mechanical prophylaxis was not captured as a structured study variable in the original protocol; therefore, the present study cannot evaluate the additive or synergistic effects of mechanical and pharmacological thromboprophylaxis. Structured patient education regarding VTE risk, correct self-injection technique, and recognition of minor injection-site reactions may contribute to successful implementation of postpartum thromboprophylaxis and treatment adherence.^22,23 Future prospective studies should incorporate standardized documentation of mechanical prophylaxis, patient education, adherence, and real-world effectiveness.

Safety Profile of LMWH

For patients indicated for pharmacological prophylaxis, selecting the appropriate agent is paramount. Study data exhibit the extensive use of LMWH, accounting for 99.1% of usage. The predominant use of LMWH is consistent with contemporary obstetric practice because of its predictable dosing and established role in peripartum anticoagulant prophylaxis. ¹⁰

The clinical outcomes from the treated cohort confirm favorable short-term inpatient tolerability of the protocol, with no severe bleeding episodes recorded during hospitalization. Adverse effects were predominantly local reactions with an incidence below 10%. The relatively low frequency of injection-site ecchymosis and pain observed in the present study should be interpreted in the context of the short inpatient follow-up period (typically 3–5 days). Delayed local reactions occurring during the remainder of the recommended postpartum thromboprophylaxis course were not captured after hospital discharge. This limitation may partially explain the lower incidence of ecchymosis and injection-site pain compared with studies evaluating the complete postpartum prophylaxis period. Similar findings have been reported in postpartum studies evaluating patient experience during extended LMWH prophylaxis, where cumulative injection-site bruising, pain, and treatment burden became more apparent over the complete post-discharge prophylaxis course.^22,23 Future studies incorporating patient-reported outcomes and adherence monitoring throughout the complete postpartum prophylaxis period may provide a more comprehensive assessment of treatment acceptability. To attain this level of safety, coordination with the anesthesiology department is essential. Strict adherence to anesthetic safety windows helps reduce complications such as spinal hematoma. ¹⁰

Thrombophilia Screening

Genetic and acquired thrombophilia may contribute to obstetric VTE, particularly among women with previous VTE, a strong family history of thrombosis, recurrent pregnancy loss, severe placenta-mediated complications, or thrombosis at unusual sites ²⁴ . Recent laboratory and hereditary thrombophilia guidance emphasizes that testing should be selective, clinically interpretable, and performed when results are likely to alter management, rather than used as universal screening in low-risk populations.^25,26 In this cohort, no patient had documented previous VTE, family history of VTE, or known high- or low-risk thrombophilia, and routine thrombophilia screening was not performed. Universal thrombophilia screening for all women undergoing cesarean delivery is therefore not supported by the present data. In Asian populations, where Factor V Leiden and prothrombin G20210A mutations are uncommon, selective testing should also consider antithrombin, protein C, protein S deficiency, and antiphospholipid antibodies according to personal history, family history, and clinical phenotype. Comtemporary evidence also supports an association between hereditary thrombophilia and venous thromboembolism.

Study Limitations

Because no symptomatic VTE events occurred during the short inpatient observation period, the present study was not designed to evaluate discrimination, calibration, sensitivity, specificity, or overall predictive performance of either risk assessment model. Only adequately powered prospective cohorts or randomized trials with complete 6-week postpartum follow-up can evaluate risk assessment models against clinical outcomes while ensuring an acceptable bleeding risk.

Additionally, this report is founded on a single-center prospective cohort design conducted at a tertiary referral hospital; hence, the epidemiological characteristics may not fully represent the national obstetric population. Several clinically relevant variables, including gestational diabetes and detailed management components of postpartum hemorrhage such as uterotonic escalation, intrauterine balloon tamponade, and intensive care admission, were not systematically captured in the original analytic protocol. Moreover, while the sample size provides reliability for analyzing risk group distributions, it limits statistical power to detect rare VTE outcomes. Multicenter prospective cohort designs with standardized 6-week follow-up are imperative to further evaluate the impact of NG compliance on thromboembolic morbidity, bleeding, adherence, and patient-reported injection burden.

Conclusion

Venous thromboembolism following cesarean section remains a significant clinical challenge, amplified by trends of advancing reproductive age and rising obesity rates. This analysis confirms that 29.0% of postpartum women after CS met criteria for pharmacological prophylaxis based on the 2023 National Guidelines, whereas the Caprini score generated prophylaxis indications for 45.3% of the cohort. The 2023 NG generated fewer prophylaxis indications than the Caprini score in this cohort, but the present study cannot determine which model more effectively prevents VTE. The prophylactic regimen utilizing LMWH demonstrated favorable short-term inpatient safety, with no clinically relevant non-major bleeding, severe hemorrhage, thrombocytopenia, DVT, or PE observed before discharge. Future multicenter prospective studies with standardized 6-week postpartum follow-up are required to evaluate obstetric risk models and determine their impact on thrombotic and bleeding outcomes.

ORCID iDs

Bui Chi Thuong https://orcid.org/0000-0001-9807-9930

Ta Viet Hung https://orcid.org/0009-0000-4503-0885

Pham Dinh Nguyen https://orcid.org/0009-0009-4611-2888

Ethical Considerations

The study protocol was approved by the Institutional Review Board (IRB) of Nhan Dan Gia Dinh Hospital (Approval Code: IRB-NDGD-2025-042).

Consent to Participate

Written informed consent was obtained from all participants prior to their inclusion in the study.