Feasibility of 2-Hour vs 24-Hour Delayed Radiography in Oil-Based Hysterosalpingography: A Comparative Diagnostic Accuracy Study

Introduction

Infertility is defined as the failure to achieve a clinical pregnancy after 12 months of regular unprotected sexual intercourse [1]. Common etiologic causes of infertility include ovulatory dysfunction, male factor infertility, and tubal disease, with tubal infertility accounting for 11% to 67% of infertility cases [2,3], depending on the population studied. Tubal infertility should be suspected in women who have any of the following risk factors: a history of sexually transmitted infections, cervical dysplasia, uterine surgery, abdominal surgery, or intra-abdominal infection [1,4]. Hysterosalpingography (HSG) involves injecting a water- or oil-soluble contrast medium through the cervix into the uterine cavity and fallopian tubes to assess their condition [5]; it has a sensitivity and specificity of 65% and 83%, respectively [6]. HSG plays an important role in assessing the reproductive function of female fallopian tubes and is the preferred method for investigating tubal patency in China [7].

Ethiodized poppyseed oil injection, an organoiodine compound that combines poppyseed oil with iodine, is an oil-soluble contrast medium. Previous studies have reported higher pregnancy rates in women using oil-soluble contrast agents than in women who previously underwent HSG using a water-soluble contrast agent [8,9]. This may be because oily media flush debris and dislodge mucus plugs from damaged tubes [10]. Therefore, more patients choose an oil-soluble contrast agent. However, patients who undergo HSG using oil-soluble contrast media require a delayed radiograph 24 hours after injection [11–13], which increases the cost to the patient in terms of time and money, consequently reducing their compliance. Some hospitals have shortened the delayed photography time to 2, 4, or 6 hours. A recent study reported that increasing the amount of exercise performed by patients after HSG (walking more than 12 000 steps) can shorten the time before delayed radiographs are obtained while ensuring diagnostic accuracy [14]; however, it still takes 1 day to complete this examination. If the delayed radiography time can be shortened to 2 hours while ensuring accuracy, patients can complete the whole examination during the morning or afternoon. For doctors, it can lead to improved diagnostic efficiency and the formulation of more timely treatment plans. For patients, it can significantly reduce costs, reduce transportation and accommodation costs to and from the hospital, and reduce anxiety and dissatisfaction caused by a long waiting time. Therefore, the promotion of this technology could be beneficial both to doctors and patients.

Laparoscopy with chromopertubation is widely accepted as the gold standard for assessing tubal patency [15]. However, its clinical application is limited by surgical risks, considerable cost, and a prolonged period of postoperative recovery. Traditionally, hysteroscopy has been a technique used to examine the uterine cavity for diagnostic purposes. It has evolved to a valuable means of simultaneously diagnosing and treating a variety of intrauterine lesions. Hidden microscopic lesions in the uterine cavity can affect the intrauterine environment and lead to poor pregnancy outcomes [16]. Hysteroscopy can identify and treat abnormalities in the reproductive tract, such as uterine adhesions, adenomyosis, T-shaped uterus, and mediastinal uterus, and it offers greater specificity for the diagnosis of endometritis [17]. In China, there are also guidelines recommending hysteroscopy for infertile patients to clarify the causes of infertility and provide targeted treatment to improve fertility [18,19]. A previous study has shown that hysteroscopic tubal assessment had a high accuracy, with the laparoscopic dye method used as reference [20,21]. In recent years, several studies have demonstrated that hysteroscopic hydrotubation is an effective method for assessing tubal patency [20,22], and that it can be used as an alternative standard for assessing tubal patency, as it is safe, inexpensive, and easy to perform. Moreover, it has a certain therapeutic effect on tubal occlusion, which can increase the fertility rate; therefore, it is widely used in clinical practice [23]. Hysteroscopy with chromopertubation (HSC) can be used to confirm patency after HSG testing and as a treatment for tubal obstruction, in which hysteroscopic-guided tubal cannulation is performed, followed by the injection of methylene blue into the fallopian orifice. In the present study, we investigated the feasibility of reducing the delayed radiography time by comparing the consistency of 2-hour and 24-hour delayed radiography during HSG for diagnosing tubal patency, using HSC results as the reference.

Material and Methods

ETHICS STATEMENT:

Prospective comparative study approval was obtained from the Medical Ethics Committee of the Women and Children’s Hospital Affiliated with Xiamen University (KY-2024-129-K01), and informed consent was obtained from all patients included in the study.

POPULATION:

Data from 515 patients who underwent HSG at the Women and Children’s Hospital Affiliated with Xiamen University, Fujian, China, from January 2019 to September 2024 were screened for the study. The inclusion and exclusion criteria for this study were formulated with reference to expert consensus [13]. The inclusion criteria were as follows: (1) age 20 to 45 years; (2) fulfilled the clinical diagnosis of infertility; (3) developmental malformation of the reproductive tract; (4) preparing for another pregnancy after an ectopic pregnancy; (5) review after tubal surgery; and (6) suspected uterine adhesions or uterine scar diverticula. The exclusion criteria were as follows: (1) history of iodine allergy, hyperthyroidism, or thyroid tumor; (2) vasculitis, sexual intercourse, or vaginal medication within 3 days prior to the HSG; (3) pelvic inflammatory disease or active pelvic, uterine, or tubal tuberculosis; (4) uterine or cervical bleeding; (5) pregnancy cannot be excluded; (6) body temperature higher than 37.5°C; and (7) refusal of hysteroscopic catheter hydrotubation after contrast surgery to assess tubal patency. After screening, 397 patients were included in the study, including 194 patients in the control group who underwent delayed radiographs 24 hours after HSG and 203 patients in the experimental group who underwent delayed radiographs 2 hours after HSG. We collected clinical information findings related to HSG and HSC, including age, body mass index (BMI), gravidity, and parity.

PROCEDURES:

HSGs were performed by gynecologists between day 7 and day 12 of the menstrual cycle. HSG images were acquired by radiologic technologists and subsequently interpreted by radiologists. Twenty-four hours after HSG examination, all films (including both the 2-hour and 24-hour delayed images) were collected. These films were anonymized, pooled together, and assigned a random study ID, to blind the radiologists to the patient’s identity and, crucially, to whether the image was taken at the 2-hour or 24-hour mark. The same team of radiologists then independently reviewed and interpreted all the randomized films. The radiologists were blinded to each other’s assessments and to the group assignment (2-hour vs 24-hour) of each film. This process ensured that the interpretation of the 24-hour films could not influence, or be influenced by, the interpretation of the 2-hour films. No premedication was used prior to HSG. Hystereography tubes (Nantong Sanli disposable hysterography tubes [F12, China]) were inserted into the cervical canal and 10 mL ethiodized poppyseed oil (Hengrui Medicine, China) was gently injected into the uterine cavity. We used an infusion pump for the administration of the contrast medium, with the pressure precisely maintained at 100 mL per hour. After the contrast agent was injected, images were recorded using the AXIOM Digital Gastrointestinal System to observe the contrast agent flowing through the uterine cavity and the fallopian tube. Then, the patients were informed of their condition by the 2 doctors in charge, informed of the benefits and risks of the 2 groups of delayed radiographs, and classified into a group according to their willingness. Delayed radiographs were obtained for the 2 groups at either 2 hours or 24 hours after HSG. The doctors instructed the patients in the 2-hour delayed radiograph group to increase their physical activity to accelerate pelvic contrast medium diffusion, and to walk at least 3000 steps within 2 hours. Using the radiographs, tubal patency was categorized as tubal patency, defined as opacification of all tubal segments with subsequent free intraperitoneal spillage; partial tubal patency, defined as opacification of all tubal segments with delayed free intraperitoneal spillage after pressurization; or tubal occlusion, defined as opacification of some tubal segments without subsequent free intraperitoneal spillage [5,24] (Figure 1).

In China, there are guidelines recommending hysteroscopy for patients with infertility [18,19]. In the present study, more than 90% patients had diagnosed infertility, the remaining patients showed suspicious signs of endometrial abnormalities during HSG, such as endometrial polyps or uterine adhesions. If intrauterine pathology, such as endometrial polyps or intrauterine adhesions that could potentially impact fertility, was incidentally discovered during a hysteroscopic examination, surgical treatment was performed concurrently. Therefore, all patients underwent hysteroscopic hydrotubation without sedation or anesthesia the following month after HSG. All hysteroscopic procedures were performed by a team of gynecologic surgeons experienced in hysteroscopy. Hysteroscopic procedures were performed 3 to 7 days after menstruation. All patients received postoperative antibiotic therapy to prevent secondary pelvic inflammatory disease. Anisodamine (Minsheng Medicine, China) was injected routinely prior to hysteroscopic incubation. Hysteroscopy uses a rigid 0° optical lens with a 5.2-mm external sheath (Shenda, Shenyang, China). A plastic catheter (KangGe, Shanghai, China) was inserted through the hysteroscopic operating channel, and its tip was inserted 1 to 3 cm into the interstitial area of the fallopian tube. Methylene blue dye (2–20 mL; methylthioninium chloride injection 10 mg, lidocaine hydrochloride injection 0.05 g, dexamethasone sodium phosphate injection 2.5 mg, gentamicin sulfate injection 80 000 units, and normal saline 20 mL) was slowly injected through the catheter. According to the resistance to the injection and the reflux of the methylene blue dye, the patency of the fallopian tubes was divided into 3 types: tubal patency, defined as no resistance to injection with flow of the methylene blue dye in the catheter, a blue catheter, and a normal-colored endometrium; partial tubal patency, defined as resistance to the injection, a blue catheter due to the flow of the methylene blue dye, and a normal-colored endometrium; and tubal occlusion, defined as the uterine cavity turning blue as a result of high resistance, causing the methylene blue dye to reflux and not be injected into the fallopian tubes (Figure 2). The procedure was performed for each fallopian tube. If hysteroscopy incidentally revealed intrauterine pathology, such as endometrial polyps or uterine adhesions, surgical treatment was performed simultaneously.

FOLLOW-UP:

All participants were monitored via telephone or email for clinical pregnancy within 12 months after the procedure (defined as an intrauterine gestational sac confirmed by ultrasonography). Concurrently, instances of abnormal pregnancies such as biochemical pregnancy, ectopic pregnancy, and missed abortion were recorded.

STATISTICAL ANALYSES:

Statistical analyses were performed using IBM SPSS (version 23.0; IBM Corp, Armonk, NY, USA) and R software (version 4.3.3). A chi-square test was used to detect differences in clinical characteristics treated as categorical variables. Furthermore, t tests were used to compare continuous data, such as BMI. We compared the 2-hour and 24-hour standard delayed radiograph results with the hysteroscopic results. The results of hysteroscopic incubation were used as a reference, and the diagnostic consistency between the 2 groups was evaluated using the kappa test. A kappa value ≥0.75 indicates high consistency. A kappa value between 0.4 and 0.75 indicates moderate consistency, and a kappa value≤0.4 indicates poor consistency. Moreover, using the outcomes of HSC as a benchmark, we assessed and compared the diagnostic performance indices of 2-hour and 24-hour delayed radiographs for distinguishing different levels of tubal patency status. Unless otherwise stated, data are presented as the mean±standard deviation (SD) or number (%). Statistical significance was defined as P<0.05.

Results

PATIENT CHARACTERISTICS:

The trial enrollment process is summarized in Figure 3. A total of 515 patients who underwent HSG were screened, 118 of whom were excluded for various reasons. Of these, 65 patients refused hysteroscopic hydrotubation, 30 patients had inflammation of the reproductive tract, 10 patients had uterine bleeding, and 13 patients had a history of intercourse during the month of HSG. Finally, 397 patients were included in this study, of whom 194 underwent delayed radiographs 24 hours after HSG (control group) and 203 underwent delayed radiographs 2 hours after HSG (experimental group). There were 367 patients with infertility and 30 patients with other indications (non-infertile). The clinical characteristics of the 397 patients are summarized in Table 1.

Overall, within 1 year postoperatively, there were 25 cases lost to follow-up in the experimental group (25/203, 12.31%), and 10 cases in the control group (10/194, 5.15%). Ultimately, a total of 126 patients in the experimental group achieved pregnancy (126/203, 62.07%), while 118 patients in the control group achieved pregnancy (118/194, 60.82%). Among them, 111 patients in the experimental group had intrauterine pregnancy (111/203, 54.68%), while 108 patients in the control group had intrauterine pregnancy (108/194, 55.67%). There was no significant difference between the 2 groups in pregnancy rates within the first postoperative year. Also, there were no significant differences between the 2 groups in terms of baseline clinical characteristics, including age, BMI, gravidity, parity, endometriosis, or the reason for HSG.

COMPARISON OF TUBAL PATENCY FINDINGS BETWEEN HYSTEROSCOPIC HYDROTUBATION AND THE 24-HOUR AND 2-HOUR DELAYED RADIOGRAPHY GROUPS AFTER HSG:

Among the 194 patients for whom radiographs were obtained 24 hours after HSG, there were a total of 384 fallopian tubes, as 4 fallopian tubes had been removed for ectopic pregnancy. Based on the radiographs, 128 tubes (128/384, 33.33%) were identified as patent, 193 (193/384, 50.26%) as partially patent, and 63 (63/384, 16.41%) as occluded (Table 2). Using the results of HSC as a reference, the concordance rate was 77.34%.

Among the 203 patients for whom radiographs were obtained 2 hours after HSG, there was a total of 400 fallopian tubes, as 6 fallopian tubes had been removed for ectopic pregnancy. HSG identified 109 tubes (109/400, 27.25%) as patent, 202 (202/400, 50.50%) as partially patent, and 89 (89/400, 22.25%) as occluded (Table 3). Using the results of HSC as a reference, the concordance rate was 76.25%.

The agreement in tubal patency findings between hysteroscopic hydrotubation and the 2 delayed radiography groups is shown in Table 4. The McNemar-Bowker test indicated a significant difference between the 2-hour and 24-hour data and the results of hysteroscopic hydrotubation (P<0.001). However, the kappa coefficients were 0.601 and 0.610, which fall within the range of 0.40 to 0.75. Diagnostic indices of 2-hour and 24-hour delayed radiographs for distinguishing different levels of tubal patency status are presented in Table 5. Using hysteroscopic hydrotubation results as the reference standard, the accuracy, precision, sensitivity, and specificity of the 3 categories were used to compare the diagnostic value of the 2 groups. When the tubes were identified as patent, the accuracy, precision, and specificity of the 2-hour and 24-hour delayed radiographs were highly comparable. The sensitivity in the control group (0.731) was slightly greater than that in the experimental group (0.679). When the tubes were identified as partially patent, the accuracy and sensitivity were highly comparable between the 2-hour and 24-hour delayed radiography groups, and the precision and specificity in the experimental group (0.931and 0.915, respectively) were greater than that in the control group (0.896 and 0.881, respectively). When the tubes were identified as occluded, the precision in the 2 groups was very low. In the experimental group, the sensitivity was greater than that in the control group, whereas the accuracy and specificity were lower in the experimental group than in the control group. Overall, the 2 groups showed only minor differences in accuracy, positive predictive value, sensitivity, and specificity, indicating very similar performance.

Discussion

MAIN FINDINGS AND CLINICAL IMPLICATIONS:

In this study, the diagnosis of tubal patency using 2- and 24-hour delayed radiography demonstrated a moderate and similar degree of consistency with the findings obtained using hysteroscopic hydrotubation. This suggests that 2-hour delayed radiography could be a viable alternative to conventional 24-hour delayed radiography. These preliminary findings warrant confirmation in a randomized controlled trial. Shortening the delayed radiography interval has the potential to significantly increase the efficiency of clinical diagnosis and may simultaneously reduce the financial and temporal burden on patients.

COMPARISON WITH EXISTING LITERATURE:

A previous study revealed that increasing the amount of exercise in patients after HSG reduced the time to delayed radiograph uptake to 6 hours while ensuring diagnostic accuracy [14]. Similar results were observed in our study. Compared with hysteroscopic hydrotubation, the concordance rate was 77.34% in the group in which delayed radiographs were obtained 24 hours after HSG, and the concordance rate was 76.25% in the group in which delayed radiographs were obtained 2 hours after HSG. Our study of the diagnostic accuracy for the assessment of tubal patency relied on several critical metrics, including the kappa coefficient, accuracy, precision, sensitivity, and specificity. The results of this comprehensive analysis indicate a moderate level of diagnostic performance for both methods. These findings suggest that using a 2-hour delayed radiograph as an alternative to the conventional 24-hour delayed radiograph for diagnosing tubal patency is a viable option. Reducing the time to delayed radiograph improves diagnostic efficiency and demonstrates significant benefits in terms of patient convenience and satisfaction. However, due to ethical review limitations, in our study, the patients who underwent 2- and 24-hour delayed radiography were groups of different individuals; because of this design, the observed similarity between groups, while promising, must be validated in a randomized setting before clinical adoption can be widely recommended.

Previous studies have demonstrated that the clinical pregnancy rate within 1 year after HSC can reach approximately 75% [23], which is consistent with our findings. Moreover, we observed no significant difference in pregnancy rate within the first postoperative year between the experimental and control groups. This suggests that the choice between 2-hour or 24-hour delayed radiography during HSG does not influence the fertility of patients. Furthermore, our research indicates that performing HSC after HSG indeed benefits the enhancement of fertility in patients.

Tubal occlusion is a prevalent cause of female infertility. HSG is recognized for its accuracy in diagnosing distal tubal occlusions [29]. However, HSC may not effectively discern whether fallopian tube obstruction is distal or proximal. Particularly for distal tubal obstructions, laparoscopic surgery offers superior diagnostic and therapeutic capabilities [30]. However, there is a high rate of false-positives for proximal tubal obstruction, due to tubal spasms and blockage of tissue fragments [31]. Tubal spasm during the procedure can constrict the tubal lumen, potentially leading to a misdiagnosis of obstruction. Additionally, the presence of tissue debris or fragments within the tubes can obstruct the flow of contrast medium, further complicating the diagnosis [32]. In our study, using hysteroscopic hydrotubation as the reference standard, the concordance rates for the diagnosis of tubal patency and partial tubal patency were higher in the control and experimental groups (86.92% and 89.07%, respectively). However, in the HSG assessment of tubal occlusion, the false-positive rates were 71.43% in the control group and 68.54% in the experimental group. In our study, shortening the delayed radiograph time did not increase the incidence of false-positive results for tubal occlusion. Although the potential influence of anesthesia – eliminating tubal spasms induced by pain – was considered, all patients in our study underwent hysteroscopic hydrotubation without anesthesia. The false-positive rate in our cohort was significantly higher than the 41.2% reported in another study [21]. This discrepancy may be attributed to the preoperative administration of anisodamine in all patients undergoing hysteroscopic hydrotubation to relieve tubal spasms, whereas no medications were administered prior to HSG. In contrast, no medications were used prior to HSG. Given the high false-positive rate for tubal occlusion, comparison of tubal patency results between hysteroscopic hydrotubation and the 2 HSG groups using the McNemar-Bowker test demonstrated significant differences between both the 2-hour and 24-hour delayed imaging groups and hysteroscopic hydrotubation. This finding aligns with existing literature, which also reports only moderate diagnostic agreement between laparoscopy and HSG, largely attributable to the high false-positive rates of HSG in detecting tubal obstruction [21]. Collectively, these observations underscore the persistent challenges in noninvasive tubal assessment. Despite this limitation, HSG remains a widely adopted initial screening tool for evaluating tubal patency, owing to several practical advantages: it is noninvasive, cost-effective, readily accessible, and provides simultaneous functional assessment of both tubal patency and uterine morphology. Additionally, HSG may offer therapeutic benefits through tubal flushing, which has been associated with enhanced fertility in some patients. While laparoscopic and hysteroscopic methods provide higher diagnostic accuracy and direct visualization, HSG continues to serve as a valuable first-line screening modality in routine infertility workups – particularly in resource-limited settings or when minimally invasive options are preferred. Nevertheless, its suboptimal positive predictive value for true anatomical obstruction necessitates a cautious interpretive approach. Therefore, in cases in which HSG suggests unilateral or bilateral tubal blockage – especially in the absence of a clinical history strongly indicative of pelvic disease – further confirmatory evaluation via laparoscopy or hysteroscopy is recommended to avoid unnecessary interventions and to guide appropriate clinical management.

STRENGTHS, LIMITATIONS, AND FUTURE RESEARCH DIRECTIONS:

We innovatively proposed shortening the delayed radiography interval after HSG to 2 hours and achieved comparable diagnostic performance, thereby potentially reducing patient costs, improving clinical efficiency, and enhancing patient satisfaction. However, patients who underwent 2-hour and 24-hour delayed radiography were not the same individuals, which may have affected the statistical results. Ideally, to provide a more objective comparison of the advantages and disadvantages of the 2 approaches, both 2-hour and 24-hour delayed radiographs should be performed in the same patients. However, due to ethical review limitations, it is not possible to perform 2 imaging procedures on the same patient in a short period of time. Other limitations are the relatively small sample size and that our study was a single-center design, which can affect the generalizability of the results. Our standardized protocols for HSG and hysteroscopy may not reflect practice variations elsewhere. While this standardization was crucial for the internal validity of our direct comparison between the 2-hour and 24-hour groups, it can limit external validity. Future multi-center studies and randomized trials are needed to confirm our findings across a wider spectrum of clinical practices and patient populations.

Conclusions

A 2-hour delayed radiograph is a viable alternative to the conventional 24-hour delayed radiograph. Shortening the delay in radiography has the potential to significantly increase the efficiency of clinical diagnosis and may simultaneously reduce the financial and temporal burden on patients. However, these are preliminary findings and require validation through randomized trials, in which each patient serves as their own control.