Nội dung

T¹P CHÝ Y - D¦îc häc qu©n sù sè 9-2020 ULTRASOUND IMAGING AND SERUM Β-HCG AS AN ADJUNCT TO DIAGNOSIS OF HYDATIDIFORM MOLE EVALUATING Β-HCG SURVEILLANCE POST-EVACUATION Trinh The Son1, Nguyen Thanh Tung1, Nguyen Ngoc Nhat1 Vu Van Tam2, Do Quang Anh2, Trinh Van Tam3, Do Nhu Binh4 SUMMARY Objectives: To evaluate characteristics of hydatidiform mole, especially ultrasound imaging and serum β-hCG level. Moreover, β-hCG monitoring strategy posthydatidiform mole gestational neoplasia was analysed. Subjects and methods: A retrospective descriptive study was performed in 244 patients with molar pregnancy. We recorded gestational age, the prevalence of vaginal bleeding, transvaginal ultrasound findings in histological confirmation. Serum β-hCG measurement before aspiration as well as post-treatment surveillance was implemented by enzyme-linked immunosorbent assay. Results: Mean gestational age was 7.92 ± 3.17 weeks. Patients with complete hydatidiform mole had significantly higher gestational age than partial hydatidiform mole (8.66 ± 3.47 weeks vs. 7.22 ± 2.68 weeks, p = 0.001). The persistence of vaginal bleeding was 49.6% and had a significant difference between complete moles and partial moles (58.3% vs 40.2%, p = 0.005, respectively). The sensitivity of transvaginal ultrasound diagnosing hydatidiform mole was 98.8% (239/242 patients). Serum βhCG level often ranged from 100,000 to 500,000 mIU/mL (52.7%), whereas complete moles had lower levels of beta-hCG in patients with serum β-hCG level less than 100,000 mIU/mL than partial moles (18.4 mIU/mL vs 44.8 mIU/mL, p < 0.05). Besides, β-hCG monitoring postevacuation was detected in 38 patients with complication requiring chemotherapy. β-hCG level post-evacuation decreased below 5 mIU/mL was 8.89 ± 3.42 weeks, and this period of time was longer than in patients with complication (12.1 ± 4.0 weeks vs 7.67 ± 2.15 weeks, p < 0.05). Conclusion: Most molar pregnancies were diagnosed in the first trimester before the onset of typical clinical symptoms due to high - resolution ultrasonography and serum β-hCG measurement adjunct. β-hCG monitoring post-evacuation was useful to detect gestational neoplasia posthydatidiform mole. * Keywords: Gestational trophoblastic disease; Hydatidiform mole; Molar pregnancy; Gestational neoplasia; Complete mole; Partial mole. 1 Military Institute of Clinical Embryology and Histology, Vietnam Military Medical University 2 Hai Phong Hospital of Obstetrics and Gynecology 3 Andrology and Fertility Hospital, Ha Noi 4 Department of Infectious Diseases, Division of Military Science, Vietnam Military Medical University Corresponding author: Do Nhu Binh (nhubinh.do@vmmu.edu.vn) Date received: 10/10/2020 Date accepted: 15/12/2020 147 T¹P CHÝ Y - D¦îc häc qu©n sù sè 9-2020 INTRODUCTION Gestational trophoblastic disease (GTS) is a group of rare diseases related to pregnancy disorder including a group of interrelated diseases ranging from the premalignant partial hydatidiform mole (PHM) and complete hydatidiform mole (CHM) to the malignant type called gestational trophoblastic neoplasia (GTN). GLN consists of an invasive mole, choriocarcinoma, placental-site trophoblastic tumor (PSTT) and epithelioid trophoblastic tumor (ETT) [1]. Hydatidiform mole is the most common pathology of GTS, associated with abnormal gametogenesis and fertilization. The prevalence of this disease is various depending on the region. In North America and Europe, the incidence of PHM and CHM is approximately three per 1,000 pregnancies. However, it appears higher in Asia countries as well as native American Indians [2]. while the highest was 202 in 100,000 Chinese pregnancies. Historically, vaginal bleeding, a uterus greater than the fetal development, preeclampsia, hyperemesis, hyperthyroidism, and respiratory distress often fully present in the second trimester. However, there has been a change in the clinical symptom of hydatidiform mole, with abnormal vaginal bleeding being the most common presentation, sometimes, it was the only one clinical presentation [3]. Therefore, at present, vaginal ultrasonography and β-hCG measurement adjunct as a useful tool to detect hydatidiform mole earlier in the first trimester. Gestational trophoblastic neoplasia is a malignant form that can develop after full-term delivery, a spontaneous miscarriage, 148 and posthydatidiform mole with 2.000-fold greater. There are 6 - 36% of patients who progress to GTN after evacuation [4]. At FIGO 2000, postmolar GTN was defined based on hCG-level increase, patient’s histology, and specific investigations. Consequently, β-hCG monitoring has played a crucial role to identify the risk of GTN. Therefore, we conducted this study: To evaluate symptoms for early diagnosis of hydatidiform mole, especially in the first trimester, and propose β-hCG kinetics as a factor of identifying posthydatidiform molar GTN. SUBJECTS AND METHODS 1. Subjects 244 patients with hydatidiform moles were followed at Hai Phong Hospital of Obstetrics and Gynecology from January 1, 2015, to December 31, 2018. 2. Methods * Design study: This is a retrospective descriptive study. * Procedure: All cases were treated for mole removal without chemotherapy adjunct, and then β-hCG monitoring was performed. We recorded gestational age, vaginal bleeding, transvaginal ultrasonography, and serum β-hCG level before aspiration. Typical ultrasonic appearances of hydatidiform mole were cystic changes of the placenta, absence of fetal part, and a deformed gestational sac. On the other hand, serum β-hCG measurement before aspiration as well as post-treatment surveillance was performed by enzymelinked immunosorbent assay. T¹P CHÝ Y - D¦îc häc qu©n sù sè 9-2020 Evacuation and curettage once or twice, ideally accomplished under vaginal ultrasound guidance, was an appropriate method for maintenance of fertility independent uterine size. Hysterectomy was an available alternative to evacuation if patients did not desire fertility because of childbearing complete. All molar pregnancies had a confirmed histological diagnosis. Subsequently, follow-up serum β-hCG level every week was crucial for early diagnosis and management of GTN. We measured the β-hCG level weekly until the concentration decreased below 5 mIU/ml, then further two specimens were evaluated at a weekly interval to confirm the normal value of the test. * Data analysis: Using IBM SPSS software for Windows version 22.0. Our data was described in the form ± SD, or frequency. Student t-test was used for continuous variables as appropriate. p value less than 0.05 was reflected statistically significant. RESULTS Table 1: Gestational age at evacuation. Gestational age (weeks) ≤ 10 CHM PHM (n, %) 72 (66.7) 11 - 15 31 (28.7) > 15 Total Total 100 (87.7) 11 (9.6) 172 (77.5) 42 (18.9) 5 (4.6) 3 (2.6) 8 (3.6) 108 (48.6) 114 (51.4) 222 (100.0) 22 patients were excluded due to not remembering the first day of the last menstrual cycle. Mean gestational age was 7.92 ± 3.17 weeks (range, 4 - 20 weeks of gestation). Most patients had gestational age less than or equal to 10 weeks (172/222 patients). The gestational age of CHM (8.66 ± 3.47 weeks) was longer than that of PHM (7.22 ± 2.68 weeks), this difference was statistically significant (p < 0.05). Table 2: Vaginal bleeding. Persistant vaginal bleeding (weeks) CHM n (%) PHM n (%) Total n (%) p <1 41 (55.4) 35 (74.4) 76 (62.8) 1-2 9 (12.2) 6 (12.8) 15 (12.4) >2 24 (32.4) 6 (12.8) 30 (24.8) 0.046 74 (58.3) 47 (40.2) 121 (49.6) 0.005 Total > 0.05 The overall percentage of vaginal bleeding was 49.6%. In CHM group, the prevalence was significantly higher than that of PHM group (58.3% vs 40.2%, p = 0.005). Most symptoms occurred within 1 week (62.8%). 149 T¹P CHÝ Y - D¦îc häc qu©n sù sè 9-2020 Table 3: Ultrasound and histological findings. Histopathological diagnosis CHM PHM Total (n, %) (n, %) (n, %) 124 (98.4) 115 (99.1) 239 (99.8) Stillbirth 1 (0.8) 1 (0.9) 2 (0.8) Uterine fibroids 1 (0.8) 0 (0.0) 1 (0.4) 126 (52.1) 116 (47.9) 242 (100.0) Ultrasound diagnosis Hydatidiform mole Total In total, 242 patients were diagnosed by histopathological examination. Of these cases, sonographic findings suggested a sensitivity of 99.8% in 239 cases of hydatidiform mole, although there was a slight increase in the sensitivity of 99.1% for PHM vs. 98.4% for CHM with no statistical significance. Confusing statements occurred in 3 patients after scanning with ultrasound (2 stillbirths and 1 uterine fibroids). Table 4: Serum βhCG concentration prior to evacuation. Hydatidiform mole βhCG level CHM PHM Total (n, %) (n, %) (n, %) < 100,000 23 (18.4) 52 (44.8) 75 (31.1) 100,000 - 500,000 71 (56.8) 56 (48.3) 127 (52.7) 500,000 - 1,000,000 22 (17.6) 5 (4.3) 27 (11.2) 1,000,000 - 1,500,000 8 (6.4) 2 (1.7) 10 (4.1) 1,500,000 - 2,000,000 1 (0.8) 1 (0.9) 2 (0.8) 125 (100.0) 116 (100.0) 241 (100.0) (mIU/mL) Total βhCG measurement was performed before removing the mole, the percentage of cases with βhCG level concentration ranging from 100,000 to 500,000 mIU/mL was the highest (52.7%). The incidence of cases with serum βhCG concentration < 100,000 mIU/mL of CHM and PHM were 18.4% and 44.8%, respectively. The difference was statistically significant (p < 0.05). 150 T¹P CHÝ Y - D¦îc häc qu©n sù sè 9-2020 Table 5: Complications after evacuation. Complication Non-complication Total Invasive mole GTN Distant metastasis Number of patients 125 28 37 1 191 Rate (%) 65.5 14.6 19.4 0.5 100.0 After the surgical evacuation, 65.5% of patients recovered who did not require chemotherapy combination. It was estimated that 34.5% of cases had needed combined chemotherapy due to post-evacuation complications. GTN was detected based on abnormal βhCG changes in the period of monitoring time in 19.4% of cases and invasive mole in 14.6% based on pathological findings of evacuated specimens. 1 case (0.5%) had lung metastasis. Table 6: Detection time of complication based on βhCG surveillance. βhCG surveillance (weeks) ≤8 9 - 10 Total GTN Distant metastasis Total (n, %) 30 1 31 (81.6) 7 0 7 (18.4) 37 1 38 (100.0) Average time to find post-abortion complications was 6 ± 2.3 weeks. Among 38 complicated cases, 5.3% were discovered in the first 2 weeks (including a case with metastatic lung), 76.3% of patients in 3 - 8 weeks and 18.4% of cases in 9 - 10 weeks. Table 7: Time of βhCG reduction below detection limit. Non-complication Complication Total (n = 125) (n = 50) (n = 175) 7.67 ± 2.15 12.1 ± 4.0 8.89 ± 3.42 Min 4 4 4 Max 15 28 28 Time (weeks) Mean p 0.000 The average time for βhCG plasma fell down to physiological index in both groups was 8.89 ± 3.42 weeks. Besides, it took a significantly longer time for βhCG blood level to return to normal in complicated posthydatidiform as compared to non-complicated patients (12.1 ± 4.0 weeks vs 7.67 ± 2.15 weeks, p < 0.005). In non-complicated cases, the time interval for reduction in βhCG concentration under 5 mIU/mL ranged from 4 to 15 weeks after evacuation while in the other cases, that time was between 4 from 28 weeks. The incidence of patients with a normal βhCG level in the first 9-week post evacuation was 79.2%. 151 T¹P CHÝ Y - D¦îc häc qu©n sù sè 9-2020 DISCUSSION The gestational age is determined from the first day of the last menstrual cycle to evacuated time. In some cases (22/244), the first day was not remembered exactly by the patients mostly in perimenopause. Because their menstrual disorders made it difficult to recognize that day. In a recent study, the mean gestational age was 7.92 ± 3.17 weeks, and most of the patients were less than 10 - week pregnancies (77.5%). The fetal age of CHM and PHM were 8.66 ± 3.47 weeks and 7.22 ± 2.68 weeks, respectively. There was no statistically significant difference (p = 0.001). In a previous study by Osamor JO et al, the gestational age of CHM was 15.2 weeks, and that of PHM was 19.8 weeks [5]. Thus, the mean fetal age of CHM, PHM in particular and patients in general in our study was lower than the previous study. This difference lies in better socioeconomic conditions, increasing awareness of patients, an earlier appointment with doctors when the patients had abnormal signs. Moreover, another explanation was that the common use of routine ultrasound in early pregnancy allowed to find hydatidiform mole before the onset of systemic manifestations. The clinical presentation of hydatidiform mole varied atypically. In last decades, hydatidiform mole presented in the second trimester of pregnancy, with signs of vaginal bleeding, excessive uterine size, hyperemesis, pre-eclampsia, hyperthyrodism, and respiratory distress. However, there has been change in the clinical symptom of hydatidiform mole, 152 with abnormal vaginal bleeding being the most common presentation, sometimes, it was the only one clinical presentation. Abnormal vaginal bleeding in the first trimester of pregnancy was a common symptom in patients with hydatidiform mole, accounting for 84 - 97%. In this study, the proportion of patients with vaginal bleeding accounted for 49.6%. Patients with vaginal bleeding in CHM was more significantly popular than in PHM (58.3% vs 40.2%, p < 0.05). This clinical presentation in our study was lower than Horn’s findings (67%) [6]. This was consistent with lower gestational age in our study because patients could visit hospital for atypical symptoms without a vaginal bleeding sign, moreover Horn’s study only included CHM pregnancies. Several investigators observed that persistence of vaginal bleeding was less common (41%) [7] but the sample size of patients with hydatidiform mole in this study was very small (25/12.190), so the assessment might not be objective. Most patients came to the hospital for treatment within 1 week since vaginal bleeding (62.8%). Besides, there was a difference in bleeding time between 2 types of hydatidiform mole, the rate of vaginal bleeding > 2 weeks accounted for 32.4% of patients in the CHM group and 12.8% of patients in the PHM group (p < 0.05). This difference was due to the fact that CHM was more common in the older age group who usually ignored the abnormal vaginal bleeding symptoms. Ultrasonography had the high sensitivity for detection of hydatidiform mole. In addition, ultrasound also allowed to T¹P CHÝ Y - D¦îc häc qu©n sù sè 9-2020 determine the follicular cyst. In total, 242 patients were diagnosed by histopathological examination. Compared with histopathological diagnosis, the overall pre-evacuation ultrasound detection rate of HM was 99.8%. Our study showed a higher sensitivity than previous findings [7, 8]. Moreover, ultrasound was unreliable to distinguish CHM from PHM. On the other hand, 3 cases were diagnosed incorrectly by ultrasound. The rate of misdiagnosis was quite small; however, we should combine clinical examination, serum βhCG measurement, and histopathology especially for high-risk patients, advanced maternal age, uterine fibroids, ovarian cyst, etc. In cases of miscarriages at early gestational age, the specimens should be performed histopathology regularly to avoid CHM after uterine curettage. Our three cases were diagnosed with HM as a result of histopathology. βhCG (human chorionic gonadotropin) was a glycoprotein produced by trophoblastic tissue. Therefore, it was the main marker in pregnancy and GTD. Indeed, hCG was a useful marker of disease progression, response, and post-surgical evacuation surveillance. Quantification of serum βhCG allowed to support the diagnosis in case of an unclear ultrasound. In our study, there were 241/244 cases of the quantitative βhCG test before the abortion. The majority of overall patients had βhCG concentrations from 100,000 to 500,000 mIU/mL (52.7%) which was 56.8% in CHM and 48.3% in PHM, the difference was statistically significant p < 0.05. Our results were similar to Shaaban et al’s findings [8]. Around 50% of patients with βhCG > 100,000 mIU/mL in CHM were higher than that in PHM (< 10%). We did not meet any cases of perforation of the uterine, post-hydatidiform mole infection. Complications requiring chemotherapy was present in 34.5%. It was similar to the conclusion that after molar evacuation 6 - 36% of patients progressed to GTN [4]. Trophoblastic tumor complications was found in 28 cases (15.4%) through histopathology results. According to FIGO 2000 criteria, complications requiring chemotherapy detected by monitoring of βhCG concentration were 38 cases (19.9%), including 1 patient (0.5%) with lung metastasis. Identifying post-evacuation complications was attatched much importance to identify risk factors for GTN. Several studies indicated the need for follow-up serum βhCG concentration after molar evacuation [4]. In our study, mean time for βhCG returning to normal in overall patients was 8.89 ± 3.42 weeks. In the group without complications, average time was 7.67 ± 2.15 weeks while in the complicated group was 12.1 ± 4.0 weeks, the difference was statistically significant (p < 0.05). In uncomplicated persistent trophoblastic disease, trophoblastic activity remained constantly high or even βhCG level rose in plasma after the evacuation of the mole. It was equivalent to the previous study which found that hCG fell to normal within 56 days. 56-day cut-off to define a sub-group at lower risk of pGTN originates from research in a small 153 T¹P CHÝ Y - D¦îc häc qu©n sù sè 9-2020 data set (4,205 pregnancies) [9]. According to our results, 79.2% of cases without complications went down to physiological βhCG levels within 9 weeks (60 days). Meanwhile in Herman’s research [10], these prevalences for PHM and CHM were 89% and 78%, respectively. of complete molar pregnancy. Obstet Gynecol 1995 Nov; 86(5):775-779. CONCLUSION 5. Osamor JO, Oluwasola AO, Adewole IF. A clinico-pathological study of complete and partial hydatidiform moles in a Nigerian population. J Obstet Gynaecol J Inst Obstet Gynaecol 2002 Jul; 22(4):423-425. Vaginal ultrasound works as an important auxiliary examination method with high sensitivity for diagnosis of hydatidiform mole, especially in the first trimester in case there are no manifestation symptoms. Human chorionic gonadotropin (βhCG) is widely used in the management of hydatidiform mole and posthydatidiform mole. Monitoring of βhCG levels after ovarian abortion helps to detect timely posthydatidiform GTN. REFERENCES 1. Ngan HYS, Seckl MJ, Berkowitz RS, et al. Update on the diagnosis and management of gestational trophoblastic disease. Int J Gynecol Obstet 2018; 143(S2):79-85. 2. Stevens FT, Katzorke N, Tempfer C, et al. Gestational trophoblastic disorders: An update in 2015. Geburtshilfe Frauenheilkd 2015 Oct; 75(10):1043-1050. 3. Soto-Wright V, Bernstein M, Goldstein DP, et al. The changing clinical presentation 154 4. Garavaglia E, Gentile C, Cavoretto P, et al. Ultrasound imaging after evacuation as an adjunct to beta-hCG monitoring in posthydatidiform molar gestational trophoblastic neoplasia. Am J Obstet Gynecol 2009 Apr; 200(4):417.e1-5. 6. Horn L-C, Kowalzik J, Bilek K, et al. Clinicopathologic characteristics and subsequent pregnancy outcome in 139 complete hydatidiform moles. Eur J Obstet Gynecol Reprod Biol 2006 Oct; 128(1-2):10-14. 7. Kirk E, Papageorghiou AT, Condous G, et al. The accuracy of first trimester ultrasound in the diagnosis of hydatidiform mole. Ultrasound Obstet Gynecol 2007; 29(1):70-75. 8. Shaaban AM, Rezvani M, Haroun RR, et al. Gestational trophoblastic disease: Clinical and imaging features. Radiogr Rev Publ Radiol Soc N Am Inc 2017 Apr; 37(2):681-700. 9. Bagshawe KD, Dent J, Webb J. Hydatidiform mole in England and Wales 1973-83. Lancet Lond Engl 1986 Sep 20; 2(8508):673-677. 10. Heřman J, Rob L, Robová H, et al. Histopathological and clinical features of molar pregnancy. Ceska Gynekol 2019; 84(6):418-424.