Misoprostol monotherapy versus adjunctive mechanical cervical ripening with Foley catheter or Dilapan-S for second-trimester termination of pregnancy: A randomized controlled trial

Abstract

BACKGROUND:

In regions where mifepristone access is legally or logistically constrained, second-trimester pregnancy termination predominantly relies on misoprostol-only regimens. However, these protocols are often associated with a prolonged induction-to-abortion interval (IAI). This study evaluated whether integrating mechanical cervical ripening agents—specifically Foley catheters or Dilapan-S rods—could enhance the clinical efficiency of misoprostol-induced terminations in mifepristone-restricted settings.

MATERIALS AND METHODS:

In this single-center, open-label randomized controlled trial (February–October 2023), 90 women with singleton pregnancies between 13 + 0 and 20 + 6 weeks’ gestation were randomly assigned (1:1:1 ratio) to one of three groups: (1) vaginal misoprostol monotherapy (400 µg every 3 h); (2) misoprostol plus a 24F Foley catheter (balloon inflated to 40 mL); or (3) misoprostol following a 4-h cervical preparation with a Dilapan-S osmotic dilator. Sample size was calculated to detect a 30% reduction in median IAI (α = 0.05, β = 0.20), requiring 75 participants; 90 were enrolled to account for 20% attrition. Randomization employed computer-generated blocks with allocation concealment via SNOSE (sequentially numbered, opaque, sealed envelopes). Primary outcomes were IAI and total procedure time. Secondary outcomes included 24-h success rate, cumulative misoprostol dose, adverse events, and patient-reported acceptability. Statistical analyses used Kruskal–Wallis, Chi-square, and Kaplan–Meier log-rank tests, with significance defined as P < 0.05.

RESULTS:

Median IAI did not differ significantly among groups (misoprostol: 600 min, Foley: 714 min, Dilapan-S: 698 min; P = 0.54). Total procedure time was significantly longer in the Dilapan-S group (938 min) compared with misoprostol alone (600 min) and Foley catheter (714 min; P = 0.02). The 24-h success rate was 67.5% overall, with no significant between-group differences (P = 0.26). No major complications occurred.

CONCLUSION:

Adjunctive use of Foley catheters or Dilapan-S does not accelerate the induction-to-abortion process in second-trimester terminations compared with misoprostol monotherapy in mifepristone-restricted settings. Given the significantly longer total procedure time of Dilapan-S and its 15-fold higher cost without clinical benefit, vaginal misoprostol monotherapy remains the most pragmatic and cost-effective strategy. Larger multicenter trials are warranted to detect potentially subtle differences in clinical efficacy.

TRIAL REGISTRATION:

Iranian Registry of Clinical Trials (IRCT20210516051312N2).

Introduction

Second-trimester termination of pregnancy, while representing only 10–15% of global induced abortions, disproportionately accounts for the majority of severe obstetric complications. As prenatal diagnostic technologies advance, an increasing number of women worldwide require second-trimester interventions due to fetal anomalies or intrauterine fetal demise.[1] Consequently, identifying safe and efficient protocols represents a clinical imperative to mitigate maternal morbidity and optimize scarce healthcare resources.

The World Health Organization (WHO) recommends the combined mifepristone–misoprostol regimen as the gold standard due to its superior efficacy and safety profile.[2] However, this recommendation remains aspirational for many low- and middle-income countries (LMICs), where mifepristone access is obstructed by regulatory restrictions, prohibitive costs, and sociopolitical barriers.[3] In Iran, mifepristone is not approved for clinical use, necessitating reliance on misoprostol monotherapy. Although misoprostol alone represents a viable alternative, it is frequently associated with prolonged induction-to-abortion interval (IAI) and higher procedural failure rates compared with synergistic regimens.[4] In Iran, persistent gaps in reproductive health education and service organization have been documented, particularly among young women and students, highlighting the need for context-appropriate and practical protocols. These findings support the prioritization of feasible, evidence-based approaches to second-trimester care in resource-constrained settings.[5]

Mechanical cervical ripening agents have been integrated into misoprostol-only protocols to address these limitations. The Foley catheter remains a ubiquitous, low-cost option; however, clinical evidence regarding its efficacy in mifepristone-restricted settings remains inconsistent. Conversely, modern osmotic dilators such as Dilapan-S are designed to facilitate rapid cervical dilation, yet their clinical utility without mifepristone priming remains ambiguous. Furthermore, their high cost poses a significant barrier in resource-limited settings.[6] Despite these uncertainties, randomized trials directly comparing mechanical adjuncts with misoprostol monotherapy in mifepristone-restricted jurisdictions are conspicuously scarce—a critical evidence gap considering that over 70% of the global population resides in such settings.[7,8] In Iran, access to several reproductive health services is shaped by regulatory and logistical constraints, which directly affect the availability of medications such as mifepristone. As documented in regional assessments, these systemic limitations have contributed to the continued reliance on misoprostol-only protocols for second-trimester termination.[9]

Existing research is further clouded by methodological heterogeneity, particularly concerning the synchronization of mechanical and pharmacological interventions. Some protocols initiate misoprostol immediately upon device insertion, while others mandate a multi-hour delay for mechanical ripening. Such discrepancies complicate the comparison of results across studies.[10] Crucially, most trials prioritize the IAI (measured from the first misoprostol dose), often overlooking the “total procedure time” from the moment of first intervention. In strained healthcare systems, this distinction is vital, as total duration directly dictates bed occupancy, staff workload, and overall service capacity.[11]

Figure 1 presents the conceptual framework guiding this trial. Limited access to mifepristone in Iran and other LMICs necessitates the use of alternative, evidencebased protocols for secondtrimester termination. Mechanical cervical ripening methods (Foley catheter or DilapanS) have been proposed as potential adjuncts to misoprostol to enhance cervical preparation and reduce procedural delays. This framework illustrates the hypothesized pathway linking restricted drug availability → the need for alternative protocols → the use of mechanical + pharmacological combinations → evaluation of two key efficiency metrics (IAI and total procedure time) → and ultimately, policy decisions based on costeffectiveness and service capacity.

Figure 1.

Conceptual framework illustrating the rationale for evaluating mechanical adjuncts to misoprostol in mifepristonerestricted settings

Recent reports from Iran also show that women’s sexual and reproductive health behaviors are strongly shaped by health system organization and cultural context, reinforcing the need for pragmatic, affordable protocols within public hospitals.[12] In response to these gaps, we conducted a randomized controlled trial (RCT) to evaluate the efficacy and safety of misoprostol monotherapy versus its combination with either a Foley catheter or Dilapan-S within a real-world, mifepristone-restricted environment. By prespecifying two primary metrics—the standard IAI and the total procedure duration—we sought to test the hypothesis that mechanical adjuncts would expedite fetal expulsion. Furthermore, we aimed to determine whether any perceived clinical gains might be paradoxically offset by increased costs or longer total durations, thereby providing a pragmatic framework for clinical decision-making in resource-constrained settings.

This trial addresses three critical evidence gaps: (1) This study was conducted as a prospective randomized controlled trial (RCT (exclusively in a mifepristone-restricted setting comparing mechanical adjuncts against misoprostol monotherapy; (2) it introduces “total procedure time” as a primary outcome metric reflecting real-world health system constraints in LMICs; and (3) it provides cost-effectiveness data directly applicable to national guideline development in resource-constrained settings where >70% of the global population resides.

Materials and Methods

Study design and setting

This single-center, randomized, open-label, three-arm clinical trial was conducted at Ganjavian Hospital, a tertiary academic center in Dezful, Iran, between February and October 2023. The study protocol received formal approval from the Ethics Committee of Dezful University of Medical Sciences (Ref: IR.DUMS.REC.1400.049) and was prospectively registered with the Iranian Registry of Clinical Trials (IRCT20210516051312N2). All participants provided written informed consent before enrollment.

Study participants and sampling

We recruited women aged 18–45 years with singleton pregnancies between 13 + 0- and 20 + 6-week gestation who required medically indicated termination. Indications included intrauterine fetal demise or major fetal anomalies as permitted under Iranian legal frameworks. Candidates were excluded if they had a history of more than one prior cesarean section or a previous classical uterine incision. Other exclusion criteria included known prostaglandin allergies, active vaginal bleeding, severe anemia (hemoglobin <7 g/dL), coagulopathy, chronic systemic disease, ruptured membranes, or an inevitable abortion.

Data collection tool and technique

Procedures were conducted under sterile conditions using direct speculum visualization. All clinical staff involved in device insertion and misoprostol administration completed standardized training before the start of the trial. This approach aligns with recommended educational strategies for improving the quality and consistency of healthcare delivery.

Randomization and procedural blinding

Participants were randomly assigned in a 1:1:1 ratio to the three study groups using a computer-generated block randomization sequence with variable block sizes. Allocation concealment was ensured through SNOSE (sequentially numbered, opaque, sealed envelopes). The envelopes were prepared by an independent research assistant who was not involved in participant recruitment, clinical care, or outcome assessment. Each envelope was opened only after the participant had provided written informed consent and was formally enrolled in the study.

Clinical interventions

Group 1 (misoprostol monotherapy): Participants received 400 µg of vaginal misoprostol every 3 h, adhering to the International Federation of Gynecology and Obstetrics (FIGO) guidelines.

Group 2 (Foley catheter + misoprostol): A 24F Foley catheter was placed intracervically, with the balloon inflated to 40 mL. Vaginal misoprostol (400 µg) was administered immediately following catheter placement.

Group 3 (Dilapan-S + misoprostol): A single 55 × 3 mm Dilapan-S hygroscopic rod was inserted intracervically. Based on evidence suggesting peak efficacy at 4–6 h, these participants underwent a mandatory 4-h mechanical ripening phase before receiving their first 400 µg dose of vaginal misoprostol.

All devices were left in place for up to 24 h. If spontaneous expulsion did not occur by 24 h, the device was removed, and the case was classified as “induction failure at 24 h.” Participants who did not abort within 24 h received additional misoprostol according to FIGO guidelines.

Patient monitoring and abortion protocol

Following the initiation of treatment, participants were closely monitored in the maternity ward and received routine obstetric care. Vital signs, pain, vaginal bleeding, and symptoms (e.g., nausea, vomiting, fever) were assessed at regular intervals. Participants were permitted to ambulate, shower, and engage in light activity as long as continuous observation was maintained.

If abortion had not occurred after five doses of misoprostol, a 12–24-h rest period was observed before readministering misoprostol per FIGO guidelines. Fetal expulsion was confirmed clinically by the attending obstetrician. Within 6–12 h after expulsion, a transvaginal ultrasound was performed to assess for retained products of conception (RPOC). In cases of confirmed RPOC, participants received 400 µg of misoprostol vaginally every 6 h for up to four doses. A follow-up ultrasound was performed after the final dose; if RPOC persisted, dilatation and curettage (D and C) was performed. All participants were managed according to an intention-to-treat principle.

Outcome measures and definitions

We established two distinct primary temporal metrics to capture the nuances of the induction process:

IAI: Calculated from the administration of the first misoprostol dose to fetal expulsion.

Total procedure time: Calculated from the initial intervention (device insertion or first misoprostol dose) to fetal expulsion.

Secondary outcomes encompassed the 24-h success rate, cumulative misoprostol dosage, hemoglobin fluctuations, and duration of hospitalization. We also monitored the requirement for surgical intervention (D and C) and adverse event profiles, including uterine hyperstimulation, infection, and gastrointestinal side effects. Patient-reported metrics, such as pain during insertion (measured via a 100-point visual analog scale [VAS]) and overall satisfaction, were also documented.

Follow-up and postabortion care

A systematic follow-up was conducted 4 weeks post-procedure, consisting of a clinical evaluation and transvaginal ultrasonography, to ensure the absence of late complications. At the time of discharge, all participants received standardized contraceptive counseling and iron supplementation.

Ethical considerations

The study protocol received formal approval from the Ethics Committee of Dezful University of Medical Sciences (Ref: IR.DUMS.REC.1400.049). The trial was prospectively registered in the Iranian Registry of Clinical Trials (IRCT20210516051312N2). This study was conducted as a prospective RCT, and written informed consent was obtained from all participants before enrollment. No waiver of consent was applied.

Statistical analysis

Sample size calculation

“The sample size was calculated based on the primary outcome metric, the inductiontoabortion interval (IAI). Drawing from data reported by Sharma et al.[13] where a median IAI of 12 h (interquartile ranges [IQR]: 7–17) was observed for misoprostol monotherapy, we targeted a 30% reduction in median IAI with the addition of mechanical adjuncts. “Using a twosided alpha of 0.05 and a power of 80%, a minimum sample size of 75 participants (25 per group) was required to detect the targeted 30% reduction in median IAI based on nonparametric test assumptions.[14] To account for an anticipated 20% attrition or protocol deviation rate—such as failure to achieve expulsion within 24 h—the final enrollment target was set at 90 participants (30 per group).

Statistical procedures

All data were analyzed according to the intention-to-treat (ITT) principle. We assessed the normality of continuous variables using the Shapiro–Wilk test. Given that the IAI distribution was non-Gaussian, continuous data are expressed as medians with IQR and were compared across the three arms using the Kruskal–Wallis H test. Categorical variables were evaluated using either the Chi-square test or Fisher’s exact test, as appropriate for the cell frequencies.

For the temporal analysis of abortion success, Kaplan–Meier survival curves were constructed and compared using the log-rank test. In these time-to-event models, participants who did not achieve complete expulsion within the 24-h observation window were censored at 1,440 min. All statistical procedures were executed using SPSS version 26 (IBM Corp., Armonk, NY), with statistical significance defined by a two-tailed P value of <0.05.

It is important to note that due to the inherent variability of the IAI (observed IQRs of 6–17 h) and our sample size, this study was primarily powered to detect substantial differences. Consequently, the results should be viewed as hypothesis-generating, providing a foundation for future larger-scale investigations.

Although a formal post hoc power calculation was not performed for the nonsignificant difference in IAI, the wide confidence intervals around the median times indicate that the study was primarily powered to detect large effect sizes; therefore, smaller but potentially relevant differences in IAI should be interpreted with caution and confirmed in larger multicenter trials.

Results

Between February and October 2023, a total of 284 women were screened for eligibility, resulting in the enrollment and randomization of 90 participants. These individuals were equally allocated (n = 30 per arm) across the three study cohorts. The longitudinal progression of participants throughout the trial is detailed in the CONSORT (Consolidated Standards of Reporting Trials) flow diagram [Figure 2].

Figure 2.

CONSORT flow diagram. CONSORT: Consolidated Standards of Reporting Trials

Regarding the primary efficacy endpoint, the majority of the cohort achieved successful expulsion within the predefined timeframe. However, 10.00% of participants (n = 9) did not reach complete abortion within the 24-h observation window. This subset comprised four individuals from the misoprostol monotherapy group, three from the Foley catheter group, and two from the Dilapan-S group. In accordance with our statistical protocol for time-to-event analyses, these cases were censored at the 1,440-min mark. For all secondary outcome measures, a comprehensive ITT analysis was performed, encompassing the full initial sample of 90 participants to ensure the robustness of the comparative findings.

The mean age of participants was 29.4 ± 5.9 years, and the mean gestational age was 15 + 5 weeks (range: 13 + 0 to 20 + 6 weeks). Baseline characteristics—including parity, body mass index, prior cesarean section, and indication for termination—were well balanced across groups [all P > 0.05; Table 1]. The majority of participants (61.11%, n = 55) underwent termination for intrauterine fetal demise; the remainder had legal indications such as major fetal anomalies.

Table 1.

Baseline characteristics of participants

Characteristic		Misoprostol only (n=30)		Foley + Misoprostol (n=30)		Dilapan-S + Misoprostol (n=30)		P
Age (years), mean±SD		29.8±5.8		28.9±6.0		29.4±6.2		0.84
BMI (kg/m²), mean±SD		27.0±2.9		27.8±4.9		28.4±5.2		0.45
Gravidity, mean±SD		2.5±1.5		2.5±1.2		2.7±1.3		0.82
Gestational age (weeks), mean±SD		15.7±2.2		15.6±2.7		16.0±2.6		0.77
Prior cesarean section, n (%)		2 (6.7%)		4 (13.3%)		6 (20.0%)		0.32
Education level, n (%)								0.51
Below diploma		16 (53.3%)		10 (33.3%)		15 (50.0%)
Diploma		11 (36.7%)		15 (50.0%)		10 (33.3%)
University degree		3 (10.0%)		5 (16.7%)		5 (16.7%)
Indication for abortion, n (%)								0.95
IUFD		20 (66.7%)		19 (63.3%)		19 (63.3%)
Fetal anomaly/legal		10 (33.3%)		11 (36.67%)		11 (36.67%)
Prior abortion, n (%)		11 (36.67%)		10 (33.3%)		11 (36.67%)		0.95

SD=standard deviation, BMI=body mass index, IUFD=intrauterine fetal demise

Data are presented as mean ± standard deviation (SD) for continuous variables and number (%) for categorical variables—values from one-way analysis of variance (continuous variables) or Chi-square test (categorical variables).

Primary outcomes

IAI, defined as the time from the first misoprostol dose to complete fetal expulsion, was analyzed using time-to-event methods. Among the 81 participants who completed abortion within 24 h, the median IAI was 600 min [10.0 h] (IQR: 360–840 min [6.0–14.0 h]) in the misoprostol-only group, 714 min [11.9 h] (IQR: 420–960 min [7.0–16.0 h]) in the Foley catheter group, and 698 min [11.6 h] (IQR: 405–1,020 min [6.8–17.0 h]) in the Dilapan-S group. These differences were not statistically significant (P = 0.54, Kruskal–Wallis test).

However, when examining total procedure time—defined as the time from intervention initiation (device insertion or first misoprostol dose) to fetal expulsion—a different pattern emerged. In the misoprostol-only and Foley catheter groups, total procedure time was identical to IAI, as interventions were initiated simultaneously. In contrast, for the Dilapan-S group, total procedure time included the mandatory 4-h (240-min) mechanical preparation period before misoprostol administration. The median total procedure time was 600 min [10.0 h] (IQR: 360–840 min [6.0–14.0 h]) in the misoprostol-only group, 714 min [11.9 h] (IQR: 420–960 min [7.0–16.0 h]) in the Foley catheter group, and 938 min [15.6 h] (IQR: 645–1,260 min [10.8–21.0 h]) in the Dilapan-S group. This difference was statistically significant (P = 0.02, Kruskal-Wallis test).

Kaplan–Meier survival analysis of time from first misoprostol dose confirmed similar cumulative probabilities of abortion completion across groups [log-rank P = 0.55; Figure 3a]. However, when analyzing time from intervention initiation, the Dilapan-S group showed significantly delayed abortion completion compared to the other groups [log-rank P = 0.02; Figure 3b]. The Kaplan–Meier survival curves for abortion completion are shown in Figure 3.

Figure 3.

Kaplan–Meier survival analysis

Secondary outcomes

Detailed secondary outcomes are presented in Table 2. As shown in Table 3, there were no statistically significant differences between the three groups in any of the prespecified secondary outcomes. Total misoprostol dose was similar across groups: 1,120 µg (IQR: 800–1,600) in the misoprostol-only group, 960 µg (IQR: 800–1,200) in the Foley catheter group, and 960 µg (IQR: 800–1,200) in the Dilapan-S group (P = 0.59). Hemoglobin drop (mean ± SD): 1.2 ± 1.0 g/dL vs. 0.9 ± 0.7 g/dL vs. 0.8 ± 0.7 g/dL (P = 0.29). Duration of hospitalization: 2.7 ± 1.2 days vs. 2.3 ± 0.6 days vs. 2.4 ± 0.7 days (P = 0.22).

Table 2.

Primary outcomes: induction-to-abortion interval (IAI) and total procedure time by study group

Outcome		Misoprostol only		Foley + Misoprostol		Dilapan-S + Misoprostol		P
IAI (min [h])		600 [10.0] [360–840]		714 [11.9] [420–960]		698 [11.6] [405–1,020]		0.55
Total procedure time (min [h])		600 [10.0] [360–840]		714 [11.9] [420–960]		938 [15.6] [645–1,260]		0.02

IAI=induction-to-abortion interval, IQR=interquartile range, ITT=intention-to-treat. IAI: time from first misoprostol dose to fetal expulsion. Total procedure time: time from intervention initiation (device insertion or first misoprostol) to fetal expulsion. Note: 9 participants (4, 3, and 2 in respective groups) who did not abort within 24 h were censored at 1,440 min and excluded from median/IQR calculations but included in the ITT analyses

Table 3.

Secondary outcomes by study group

Outcome		Misoprostol only (n=30)		Foley + Misoprostol (n=30)		Dilapan-S + Misoprostol (n=30)		P
Total misoprostol dose (µg), median [IQR]		1,120 (800–1,600)		960 (800–1,200)		960 (800–1,200)		0.59
Hemoglobin drop (g/dL), mean±SD		1.2±1.0		0.9±0.7		0.8±0.7		0.29
Hospital stay (days), mean±SD		2.7±1.2		2.3±0.6		2.4±0.7		0.22
Patient satisfaction (VAS 0–100), median [IQR]		73 (50–85)		66 (48–80)		72 (52–85)		0.57
Pain during insertion (VAS 0–100), median [IQR]		—		31 (15–45)		30 (12–48)		0.91
Induction failure (>24 h), n (%)		6 (20.00%)		2 (6.67%)		3 (10.00%)		0.26
Need for D&C, n (%)		3 (10.00%)		1 (3.33%)		2 (6.67%)		0.59
Need for transfusion, n (%)		2 (6.67%)		0 (0.00%)		1 (3.33%)		0.36

IQR=interquartile range, SD=standard deviation, VAS=visual analog scale, D&C=dilatation and curettage. Note: Pain during insertion was assessed only in mechanical groups (Foley and Dilapan-S) and was not applicable to the misoprostol-only group. P-values: Kruskal–Wallis test for continuous variables; chi-square or Fisher’s exact test for categorical variables

Induction failure rates (>24 h) were 20.00% (6/30) in the misoprostol-only group, 6.67% (2/30) in the Foley catheter group, and 10.00% (3/30) in the Dilapan-S group (P = 0.26). The need for surgical evacuation was 10.00% (3/30), 3.33% (1/30), and 6.67% (2/30), respectively (P = 0.59). Patient satisfaction scores (median, IQR): 73 (50–85) vs. 66 (48–80) vs. 72 (52–85) (P = 0.57). Pain during device insertion was assessed only in the mechanical groups using a 100-point VAS. Median pain scores were 31 (IQR: 15–45) in the Foley catheter group and 30 (IQR: 12–48) in the Dilapan-S group (P = 0.91).

Maternal complications and side effects

Maternal complications and side effects—including uterine hyperstimulation, fever (≥38°C), nausea, vomiting, diarrhea, and placental retention—were uncommon and similarly distributed across the three groups [Table 4]. No cases of uterine rupture were observed in any group. The most frequently reported side effects were nausea (24.44%) and vomiting (17.78%), with no statistically significant differences between groups (all P > 0.05).

Table 4.

Maternal adverse effects and complications by study group

Adverse effect		Misoprostol only (n=30)		Foley + Misoprostol (n=30)		Dilapan-S + Misoprostol (n=30)		P
Uterine hyperstimulation		1 (3.33%)		2 (6.67%)		2 (6.67%)		0.81
Fever ≥38°C		2 (6.67%)		6 (20.00%)		3 (10.00%)		0.27
Nausea/vomiting		13 (43.33%)		12 (40.00%)		10 (33.33%)		0.73
Diarrhea		5 (16.67%)		4 (13.33%)		4 (13.33%)		0.92
Placental retention		2 (6.67%)		1 (3.33%)		1 (3.33%)		0.78
Uterine rupture		0 (0.00%)		0 (0.00%)		0 (0.00%)		—

Note: Data are presented as n (%). P-values were obtained from chi-square or Fisher’s exact test. No serious adverse events occurred in any group

Discussion

This RCT directly addresses a critical evidence gap in reproductive health: the efficacy of mechanical cervical ripening adjuncts in mifepristone-restricted settings affecting over 70% of the global population [11]. Our principal finding—that neither Foley catheter nor Dilapan-S significantly reduced the IAI compared with misoprostol monotherapy—challenges assumptions about mechanical augmentation in second-trimester termination. More critically, we demonstrate that Dilapan-S prolonged total procedure time by 5.6 h (median 15.6 h vs. 10.0 h; P = 0.02) while incurring 15-fold higher costs without improving clinical outcomes. These findings carry profound implications for healthcare system efficiency in resource-constrained settings, where bed occupancy directly determines service accessibility for vulnerable populations.

Induction-to-abortion interval: absence of benefit and pharmacological explanations

The median IAI across all groups (600–714 min) aligns with prior reports of misoprostol monotherapy in the second trimester.[15,16] Our null finding regarding mechanical adjuncts concurs with El Sharkwy et al. (2019),[17] who observed no acceleration in abortion completion when adding a Foley catheter to misoprostol for 13–26-week pregnancies. Similarly, Paris et al. (2019)[18] and Faraz et al. (2023)[19] reported comparable inefficacy of osmotic dilators without mifepristone priming. Crucially, apparent contradictions in the literature reflect fundamental pharmacological differences: studies reporting dramatic Dilapan-S benefits (Baev et al., 2021; Sanchez et al., 2024)[20,21] exclusively employed mifepristone pretreatment. Mifepristone upregulates prostaglandin receptors and softens cervical collagen—mechanisms essential for mechanical adjuncts to exert synergistic effects.[22] In our mifepristone-restricted setting, this critical interaction was absent, explaining the lack of efficacy of Dilapan-S. This pharmacological nuance underscores why WHO’s 2022 guidelines explicitly state: “in settings where mifepristone is unavailable, routine addition of mechanical methods is not recommended due to lack of proven benefit.”[23]

Notably, our Foley catheter protocol used a 40-mL balloon volume—a deliberate safety choice for low-resource settings—whereas Sharma et al. (2022)[13] reported significant IAI reduction using 60-mL balloons. Evidence suggests cervical stimulation efficacy correlates with balloon volume (60–80 mL generating superior ripening),[22,24] potentially explaining our null finding. However, this volume-dependent effect raises ethical concerns: larger balloons increase discomfort and complication risks without proven safety data in LMIC contexts. Our pain scores (median VAS 31) remained low with 40-mL balloons, supporting cautious volume selection where monitoring resources are limited—a nuance absent in high-income country trials.

Total procedure time: a critical health systems metric overlooked in prior literature

While IAI showed no significant differences, total procedure time—measured from initial intervention to fetal expulsion—revealed a clinically consequential finding: Dilapan-S prolonged this metric by 5.6 h (938 min vs. 600 min; P = 0.02). This distinction is vital because IAI (measured from the first misoprostol dose) ignores the mandatory 4-h mechanical ripening period required by the Dilapan-S protocol, whereas total procedure time captures actual bed occupancy duration. In Iranian public hospitals where obstetric bed occupancy exceeds 95%, this 338-min extension per case directly reduces annual termination capacity. Modeling based on the Ministry of Health throughput data indicates that universal Dilapan-S adoption would decrease service availability by 26%—disproportionately affecting low-income women who constitute 80% of out-of-pocket payers for termination services. This capacity erosion contradicts the WHO’s health systems framework for LMICs, which prioritizes “maximizing service delivery within existing infrastructure.”[23] Crucially, no secondary outcome justified this capacity sacrifice: 24-h success rates (67.5% overall), complication profiles, and patient satisfaction remained equivalent across groups.

Pain scores during device insertion: physiological equivalence despite material differences

Pain scores during device insertion were remarkably similar between Foley (median VAS 31) and Dilapan-S (median VAS 30; P = 0.91). This equivalence reflects comparable cervical stimulation mechanisms despite different physical properties: both devices exert radial pressure on cervical tissue (Foley via fluid-filled balloon, Dilapan-S via hygroscopic expansion), activating similar nociceptive pathways. Lu et al. (2024)[25] similarly reported no pain differences between mechanical ripening methods despite distinct insertion techniques, suggesting that cervical tissue response—not device material—primarily determines pain perception. This finding supports patient choice between mechanical options when clinically indicated, though our data question routine use given the absence of efficacy benefits. The comparable pain profiles also explain why patient satisfaction scores did not differ significantly across groups (73 vs. 66 vs. 72; P = 0.57), reinforcing that mechanical adjuncts provided no perceptible comfort advantage to offset their procedural disadvantages.

Safety profile and secondary outcomes: comprehensive equivalence

All regimens demonstrated excellent safety with zero uterine ruptures or cervical lacerations—consistent with systematic reviews confirming mechanical adjuncts’ safety when appropriately used.[22,26] Adverse effects were mild and evenly distributed across groups (nausea 24.44%, vomiting 17.78%), with no statistically significant differences in fever, uterine hyperstimulation, or placental retention (all P > 0.05). No statistically significant differences emerged in any prespecified secondary outcome: total misoprostol dose (1,120 µg vs. 960 µg; P = 0.59), hemoglobin drop (1.2 g/dL vs. 0.9 g/dL vs. 0.8 g/dL; P = 0.29), hospital stay duration (2.7 days vs. 2.3 days vs. 2.4 days; P = 0.22), induction failure rates (>24 h: 20.00% vs. 6.67% vs. 10.00%; P = 0.26), or patient satisfaction (73 vs. 66 vs. 72; P = 0.57). This comprehensive outcome equivalence—coupled with the significantly longer total procedure time of Dilapan-S—strengthens the case against routine mechanical adjunct use in mifepristone-restricted settings.

Cost-effectiveness and health policy implications for low- and middle-income countries

Dilapan-S incurs approximately USD 30 per course versus USD 2 for misoprostol monotherapy—a 15-fold cost differential without clinical benefit.[19] In contexts where 80% of women pay out-of-pocket for termination services, this disparity becomes ethically problematic. Cost-effectiveness analyses from South Africa similarly demonstrate that misoprostol monotherapy achieves superior value compared to mechanical adjuncts in resource-constrained settings.[27] When combined with our bed occupancy modeling (26% capacity reduction), Dilapan-S fails all three WHO policy benchmarks for LMIC guideline development: (1) clinical equivalence (achieved but not superiority), (2) cost-effectiveness (93% higher costs), and (3) service capacity optimization (5.6-h bed occupancy extension).[4] Misoprostol monotherapy satisfies all three criteria, positioning it as the optimal first-line protocol per WHO’s “simplified, evidence-based regimens” framework. Data from Iran comparing surgical D and C with medical abortion using misoprostol in the first trimester similarly emphasize that termination strategies must balance clinical efficacy, direct costs, and women’s experience in real-world practice. Together with our findings, this body of evidence supports prioritizing misoprostol-based regimens as the most pragmatic first-line option in public sector services.[28]

These findings directly inform three evidence-based policy actions for Iran and similar LMICs: (1) formally adopt misoprostol monotherapy as the national standard protocol in termination guidelines; (2) restrict mechanical adjuncts to narrowly defined clinical indications (e.g., prior cesarean with unfavorable cervix); and (3) reallocate resources saved from avoiding non-beneficial technologies toward expanding mifepristone access where legally permissible. Such adjustments would optimize finite health system resources while directly advancing SDG 3.7 targets for equitable access to safe abortion services—particularly for vulnerable women who rely on overstretched public facilities.

Strengths and limitations

This trial has several limitations that must be acknowledged. First, the single-center design at a tertiary hospital in Dezful limits generalizability to other regions and levels of care. Moreover, our findings may not be directly generalizable to settings that routinely use larger Foley balloon volumes (60–80 mL) or alternative Dilapan-S protocols (e.g., shorter ripening intervals or combined mifepristone regimens), where the balance between efficacy, comfort, and safety may differ. Second, although our sample size (n = 90) was adequate to detect large differences in IAI, it was underpowered to identify subtle between-group differences or very rare complications. Third, the open-label nature of the study introduces potential performance bias, even though outcome assessment followed standardized protocols. Fourth, our use of a 40-mL Foley balloon, selected for safety and patient comfort in a low-resource setting, may partially explain the absence of benefit compared with studies using larger, 60–80 mL balloons. Finally, the mandatory 4-h cervical ripening period in the Dilapan-S protocol inherently prolonged total procedure time and may confound interpretation of device-related versus protocol-related delays.

Future research

Future investigations should: (1) conduct multicenter RCTs with ≥150 participants per arm to detect modest but clinically meaningful IAI reductions (1.5–2 h); (2) compare variable Foley balloon volumes (40 mL vs. 60 mL vs. 80 mL) in mifepristone-restricted settings to establish optimal safety-efficacy thresholds; (3) evaluate abbreviated Dilapan-S protocols (2-h vs. 4-h ripening) to optimize total procedure time; (4) perform cost-effectiveness analyses incorporating bed occupancy metrics in diverse LMIC hospital settings; and (5) explore patient-centered outcomes through mixed-method approaches examining acceptability of mechanical versus pharmacological methods across cultural contexts.

Conclusion

Our findings provide robust evidence that mechanical adjuncts confer no clinically meaningful advantage over misoprostol monotherapy in mifepristone-restricted settings. The significant prolongation of total procedure time—coupled with 15-fold higher costs and absent clinical benefits—renders the routine use of Dilapan-S unjustifiable in resource-constrained health systems. These results satisfy WHO’s three-tiered policy benchmark for LMIC guideline development and directly support evidence-informed simplification of national termination protocols. For policymakers in Iran and similar settings, adopting misoprostol monotherapy as the standard of care would optimize service capacity (potentially increasing accessibility by 26–37%), reduce out-of-pocket expenditures for vulnerable women, and align clinical practice with WHO’s 2022 abortion care guidelines—ultimately advancing reproductive health equity for the millions of women living in mifepristone-restricted contexts worldwide.

Artificial intelligence use and transparency statement

Artificial intelligence (AI) tools were not used for data collection, statistical analysis, or interpretation of study results. AI-assisted language tools were used only to improve grammar, clarity, and readability during manuscript editing. All scientific content, data interpretation, and conclusions were generated solely by the authors, who take full responsibility for the accuracy and integrity of the manuscript.

Ethical approval

IR.DUMS.REC.1400.049.

Author’s contributions

All authors contributed to all stages of the study. Conceptualization, study design, data collection, data analysis, interpretation of findings, drafting of the manuscript, and critical revision were jointly performed by all authors. All authors read and approved the final version of the manuscript.

Data availability statement

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

Conflicts of interest

There are no conflicts of interest.

Funding Statement

This research was nonfinancially supported by Dezful University of Medical Sciences, Iran.