Percentage of human chorionic gonadotropin change in the forty-eight hours prior to methotrexate injection in predicting treatment success

Abstract

AimAlthough many reports that examine the concentration of β-hCG are available, there is a limited number of articles in the literature about the dynamics of β-hCG. Our aim in this study was to reveal the role of β-hCG percentage change before methotrexate injection on predicting treatment success.MethodsThis retrospective study was conducted between May 2015 and February 2019 at the gynecology department of a tertiary hospital. Medical data of patients who were diagnosed with tubal ectopic pregnancy were reviewed (n:1073). The percentages of β-hCG change between 48 hours before methotrexate injection and day 1 were compared between the failure and success groups. The ROC curve was designed to determine the optimal β-hCG change percentage to predict treatment success.ResultsFour hundred and thirty-four patients were eligible for methotrexate treatment. The median value of β-hCG percentage change before methotrexate injection was significantly higher in the failure group ( + 13.7( + 9.04/ + 17.68)) compared to the success group ( + 8.62( + 5.8/ + 11.5)) (p<0.001). The area under the curve was 0.727 with 95% CI(0,659-0,795) (p<0,01). With the cutoff value of 11.27%, sensitivity, specificity, PPV and NPV were 71,2%, 73,4%, 35% and 93%, respectively.ConclusionAlong with HCG concentrations, HCG dynamics should also be investigated as to whether it has a role in the prediction of methotrexate success. We think that utilizing β-hCG dynamics as an additional tool in the prediction of methotrexate treatment outcome should be considered in every case of tubal ectopic pregnancy.

Keywords

Introduction

Ectopic pregnancy is encountered in 1-2% of all pregnancies and is accepted as a life-threatening emergency.¹ It is one of the most dangerous causes of maternal mortality in the first trimester of pregnancy.² Transvaginal ultrasound, along with serial beta-human chorionic gonadotropin (β-hCG) measurements and a high index of suspicion, is the most used tool to detect ectopic pregnancy to prevent adverse consequences.
The goal is early treatment to prevent tubal rupture, internal hemorrhage, and maternal morbidity. Methotrexate treatment is considered a feasible solution in meticulously selected cases.³ Stovall’s “single dose protocol” is the most commonly preferred among methotrexate treatment protocols.⁴
Treatment failure risk should be taken into account and cases must be selected appropriately. Presence of fetal cardiac activity, free peritoneal blood, size (> 4 cm) of the ectopic mass, high (> 5,000 mIU/mL) initial β-hCG concentration, increasing β-hCG concentrations (> 50% in 48 hours) before methotrexate, and rapid rise of hCG concentrations after methotrexate are predictive factors for methotrexate failure.^5-6-7 The documented success rate of treatment in the “single dose protocol” is between 75.45% and 87%.^8,9 Serial β-hCG measurements before methotrexate treatment are efficient to discriminate ongoing ectopic pregnancies from spontaneously resolving ones.¹⁰ Successive β-hCG measurements may detect spontaneous resolution in more than 60% of cases.¹¹
Since β-hCG rise before methotrexate treatment might indicate trophoblastic tissue activity, we tried to delineate whether β-hCG percentage change can predict methotrexate success or not. The role of β-hCG levels on day 1 and 48 hours prior to methotrexate injection in predicting success was the secondary outcome. Finally, determining a percentage change cutoff value with optimal sensitivity and specificity to predict treatment success was our tertiary outcome.

Materials and Methods

This retrospective study was conducted in accordance with the Principles of the Declaration of Helsinki and approved by the local ethics committee (Date: 2019-03-19, No: 1197). Between May 2015 and February 2019, n = 434 tubal ectopic pregnancy cases who attended our gynecology department were included.
Medical data of women diagnosed with tubal ectopic pregnancy were reviewed (n = 1073). Age, gravidity, parity, date of the last normal menstrual period, BMI, day 1 β-hCG, and β-hCG at 48 hours prior to methotrexate injection of all eligible cases were recorded. Diagnosis was established on both β-hCG level measurements and transvaginal ultrasound indicating ectopic pregnancy via an inhomogeneous mass next to the ovary, an extraovarian hyperechoic mass, or an extrauterine gestational sac with a thin endometrium.¹²
Patients presenting with hemodynamic instability, β-hCG > 10,000 IU, fetal pole with cardiac activity, severe abdominal pain, or signs of intra-abdominal haemorrhage were treated surgically (n = 238) and excluded. Cases showing spontaneous resolution (n = 370) and 31 subjects with insufficient data were also excluded. As a result, n = 434 cases were eligible. Patients were divided into two groups according to methotrexate treatment outcome: success group (n = 361) and failure group (n = 73).
Single dose protocol was used in all cases at a dose of 50 mg/m².¹³ Methotrexate administration day was defined as day 1, and β-hCG measurements were repeated on days 4 and 7. Treatment failure was defined as symptoms of tubal rupture or β-hCG not falling more than 15% between days 4 and 7. Treatment success was defined as recovery.
The aim was to reveal the role of β-hCG percentage change before methotrexate injection in predicting treatment success. The percentages of β-hCG change between 48 hours before methotrexate injection and day 1 were compared between the failure and success groups. The equation of β-hCG percentage change was as follows:
β-hCG change (%) = (β-hCG day 1 – β-hCG 48 hours prior to injection) / (β-hCG 48 hours prior to injection) × 100
Additionally, success and failure groups were compared regarding both β-hCG level on day 1 and β-hCG level 48 hours before methotrexate injection. A receiver-operating characteristic curve was designed to determine the optimal β-hCG level change for treatment success.Ethical ApprovalEthics Committee approval for the study was obtained.Statistical AnalysisAll analyses were performed using SPSS v 25.0 (SPSS Inc., Chicago, IL, USA). The Kolmogorov-Smirnov test was used to evaluate normal distribution. Descriptive statistics included frequency, percentage, mean ± SD, and median (25th–75th percentiles). The Chi-square test was used for categorical variables. Student t-test was applied for normally distributed data, while the Mann-Whitney U test was applied for non-normal data. A receiver-operating characteristic curve was utilized to establish the optimal cutoff value of β-hCG change with the highest sensitivity and specificity for treatment success. p<0.05 was considered significant.

Results

In our study, there were n = 1073 subjects diagnosed with tubal ectopic pregnancy. Of these, n = 434 patients (40%) were eligible for methotrexate treatment after excluding subjects who underwent immediate surgery (n = 238), those that resolved spontaneously (n = 370), and those with insufficient data (n = 31) (Figure 1). Characteristics of variables in the treatment success and treatment failure groups are depicted in Table 1. Differences between the two groups regarding age, gravidity, parity, BMI, and gestational week did not reach statistical significance. There was no significant statistical difference regarding day 1 β-hCG levels and β-hCG levels 48 hours before methotrexate injection between the groups (Table 2).
The percentages of β-hCG change between 48 hours before methotrexate injection and day 1 were compared between the failure and success groups. The median value of β-hCG percentage change before methotrexate injection was significantly higher in the failure group ( + 13.7 ( + 9.04/ + 17.68)) compared to the success group ( + 8.62 ( + 5.8/ + 11.5)) (p<0.001).
A receiver-operating characteristic curve was established to determine the optimal β-hCG percentage change for treatment outcome. The area under the curve was 0.727 with 95% CI (0.659-0.795) (p<0.01) (Figure 2). With the cutoff value of 11.27%, the characteristics of the model to predict treatment success were as follows: sensitivity: 71.2%, specificity: 73.4%, diagnostic accuracy: 73%, negative predictive value (NPV): 93%, and positive predictive value (PPV): 35%.

Discussion

If patients diagnosed with tubal ectopic pregnancy are not eligible for expectant management, medical management is a plausible alternative to surgery. Methotrexate therapy is the most used pharmacologic treatment for hemodynamically stable patients due to its safety and efficacy.¹⁴ Selecting appropriate patients prior to medical intervention is crucial, since compliance with follow-up is essential and treatment failure risk is always present.¹⁵ Methotrexate success for the single dose protocol is reported as a > 15% decline in β-hCG levels between days 4 and 7 following injection, with a PPV of 93%.¹⁶ Furthermore, this decline was reported as 88-100% indicating treatment success.^17,18 Although β-hCG patterns for ongoing intrauterine pregnancies and resolving pregnancies are well understood, there is no tool or model that can specify a β-hCG curve for ectopic pregnancies.¹⁹ Several studies have researched β-hCG as a marker for trophoblastic cell activity. Ferreira et al. demonstrated a direct correlation between increased serum β-hCG levels during the 48 hours prior to surgery with higher trophoblastic cell proliferation and angiogenesis in tubal pregnancy.²⁰ Oktay et al. assessed salpingectomy specimens of tubal ectopic pregnancies for myosalpingeal invasion and found higher serum β-hCG levels in patients with muscular layer trophoblastic invasion compared to those without invasion (13,665 ± 2986 mIU/mL vs. 2169 ± 870 mIU/mL; p = 0.0001).²¹ A serum β-hCG cutoff level of 5400 mIU/mL or higher had an 89% PPV and 94% NPV for detection of invasion.²¹ Pulatoglu et al. indicated that the cutoff β-hCG value detecting methotrexate failure with 71.8% sensitivity and 68.2% specificity was 1362 mIU/mL. In cases with β-hCG > 1362 mIU/mL, the failure rate was 23.9%, while at lower values it was 17.9%, which did not reach statistical significance.²² In our study, there was no difference regarding day 1 β-hCG levels and β-hCG values 48 hours before methotrexate injection between success and failure groups. Along with HCG concentrations, HCG dynamics should also be investigated for their role in predicting methotrexate success. Dudley et al. reported a significant increase in β-hCG change both before and after methotrexate injection in ectopic pregnancies that resulted in rupture: + 0.94 vs. + 0.16 and + 0.38 vs. - 0.21 (p<0.01).²³ In their 401-patient retrospective study, Cohen et al. reported that β-hCG percentage change in the 48 hours preceding MTX injection and β-hCG concentration at day 1 were independent predictors for tubal rupture (OR = 1.08, 95% CI = 1.04-1.12, p<0.001 for every percentage change in β-hCG; OR = 1.001, 95% CI = 1.0003-1.002 for every unit change in β-hCG).²⁴ They concluded that in women with β-hCG percentage increment > 69% in the 48 hours prior to injection, the probability for tubal rupture was 85%.²⁴ They also indicated that the absolute risk for tubal rupture in women with β-hCG increment < 20% is low.²⁴ In another study, da Costa Soares et al. investigated β-hCG increment before methotrexate injection as a success predictor.²⁵ The mean β-hCG increment in the methotrexate failure group was significantly higher than in the success group (36% vs. 13%). ROC curve analysis revealed the optimal cut-off value for success as 11.1%.²⁵ Similarly, in our study, the median β-hCG percentage change before methotrexate injection was significantly higher in the failure group (+ 13.7 ( + 9.04 / + 17.68 )) compared to the success group (+ 8.62 ( + 5.8 / + 11.5 )). ROC curve analysis showed the optimal cut-off value for success as 11.27% with a sensitivity of 71.2%, comparable to da Costa Soares.²⁵ We obtained a PPV of 35%, meaning a woman with a positive result in favor of failure has a 35% chance of failing treatment. On the other hand, we obtained an NPV of 93%, meaning the proportion of cases predicted as methotrexate success who were truly in the success group was 93%. Thus, our model with an 11.27% cutoff is stronger in predicting success than failure. We think that utilizing β-hCG dynamics as an additional tool in predicting methotrexate treatment outcome should be considered in every case of tubal ectopic pregnancy.

Limitations

A limitation of this study is its retrospective nature. Prospective studies should be conducted to predict methotrexate treatment outcome. In addition, during the study period decisions regarding intervention were based on clinical judgment by different physicians.

Conclusion

In conclusion, β-hCG increment before methotrexate injection is a valuable predictor for methotrexate treatment success. Patients with β-hCG increment less than 11,27% can be reassured for treatment success with an NPV of 93% irrespective of serum β-hCG concentration.

References

1. Panelli DM, Phillips CH, Brady PC. Incidence, diagnosis and management of tubal and nontubal ectopic pregnancies: a review. Fertil Res Pract. 2015;1:15. doi:10.1186/s40738-015-0008-z
   Article · PubMed · Google Scholar
2. Creanga AA, Shapiro-Mendoza CK, Bish CL, Zane S, Berg CJ, Callaghan WM. Trends in ectopic pregnancy mortality in the United States: 1980-2007. Obstet Gynecol. 2011;117(4):837-843. doi:10.1097/aog.0b013e3182113c10
   Article · PubMed · Google Scholar
3. Alur-Gupta S, Cooney LG, Senapati S, Sammel MD, Barnhart KT. Two-dose versus single-dose methotrexate for treatment of ectopic pregnancy: a meta-analysis. Am J Obstet Gynecol. 2019;221(2):95-108.e2. doi:10.1016/j.ajog.2019.01.002
   Article · PubMed · Google Scholar
4. Kanmaz AGK, Hamdi İnan A, Beyan E, Budak A. Various β-hCG monitoring protocols for predicting the success of single-dose methotrexate in the treatment of ectopic pregnancy. J Obstet Gynaecol. 2019;39(6):811-815. doi:10.1080/01443615.2019.1575344
   Article · PubMed · Google Scholar
5. Bonin L, Pedreiro C, Moret S, Chene G, Gaucherand P, Lamblin G. Predictive factors for the methotrexate treatment outcome in ectopic pregnancy: a comparative study of 400 cases. Eur J Obstet Gynecol Reprod Biol. 2017;208:23-30. doi:10.1016/j.ejogrb.2016.11.016
   Article · PubMed · Google Scholar
6. Pulatoglu C, Dogan O, Basbug A, Kaya AE, Yildiz A, Temizkan O. Predictive factors of methotrexate treatment success in ectopic pregnancy: a single-center tertiary study. North Clin Istanb. 2018;5(3):227-231.
   PubMed · Google Scholar
7. Rabischong B, Tran X, Sleiman AA, et al. Predictive factors of failure in management of ectopic pregnancy with single-dose methotrexate: a general population-based analysis from the Auvergne Register, France. Fertil Steril. 2011;95(1):401-404.e1. doi:10.1016/j.fertnstert.2010.08.025
   Article · PubMed · Google Scholar
8. Sindiani AM, Alshdaifat E, Obeidat B, Obeidat R, Rawashdeh H, Yaseen H. The use of single-dose methotrexate in the management of ectopic pregnancy and pregnancy of unknown location: 10 years' experience in a tertiary center. Int J Womens Health. 2020;12:1233-1239. doi:10.2147/ijwh.s279426
   Article · PubMed · Google Scholar
9. Gungorduk K, Asicioglu O, Yildirim G, Gungorduk OC, Besimoglu B, Ark C. Comparison of single-dose and two-dose methotrexate protocols for the treatment of unruptured ectopic pregnancy. J Obstet Gynaecol. 2011;31(4):330-334. doi:10.3109/01443615.2011.560301
   Article · PubMed · Google Scholar
10. Cohen A, Zakar L, Gil Y, et al. Methotrexate success rates in progressing ectopic pregnancies: a reappraisal. Am J Obstet Gynecol. 2014;211(2):128.e1-128.e5. doi:10.1016/j.ajog.2014.03.043
    Article · PubMed · Google Scholar
11. Levin I, Tsafrir Z, Sa'ar N, et al. Watchful waiting in ectopic pregnancies: a balance between reduced success rates and less methotrexate. Fertil Steril. 2011;95(3):1159-1160. doi:10.1016/j.fertnstert.2010.10.014
    Article · PubMed · Google Scholar
12. van Mello NM, Mol F, Ankum WM, Mol BW, van der Veen F, Hajenius PJ. Ectopic pregnancy: how the diagnostic and therapeutic management has changed. Fertil Steril. 2012;98(5):1066-1073. doi:10.1016/j.fertnstert.2012.09.040
    Article · PubMed · Google Scholar
13. Practice Committee of American Society for Reproductive Medicine. Medical treatment of ectopic pregnancy: a committee opinion. Fertil Steril. 2013;100(3):638-644. doi:10.1016/j.fertnstert.2013.06.013
    Article · PubMed · Google Scholar
14. Mohammad NS, Nazli R, Zafar H, Fatima S. Effects of lipid based multiple micronutrients supplement on the birth outcome of underweight pre-eclamptic women: a randomized clinical trial. Pak J Med Sci. 2022;38(1):219-226.
    PubMed · Google Scholar
15. Sivalingam VN, Duncan WC, Kirk E, Shephard LA, Horne AW. Diagnosis and management of ectopic pregnancy. J Fam Plann Reprod Health Care. 2011;37(4):231-240. doi:10.1136/jfprhc-2011-0073
    Article · PubMed · Google Scholar
16. Kirk E, Condous G, Van Calster B, et al. A validation of the most commonly used protocol to predict the success of single-dose methotrexate in the treatment of ectopic pregnancy. Hum Reprod. 2007;22(3):858-863. doi:10.1093/humrep/del433
    Article · PubMed · Google Scholar
17. Skubisz MM, Li J, Wallace EM, Tong S. Decline in β-hCG levels between days 0 and 4 after a single dose of methotrexate for ectopic pregnancy predicts treatment success: a retrospective cohort study. BJOG. 2011;118(13):1665-1668. doi:10.1111/j.1471-0528.2011.03133.x
    Article · PubMed · Google Scholar
18. Nguyen Q, Kapitz M, Downes K, Silva C. Are early human chorionic gonadotropin levels after methotrexate therapy a predictor of response in ectopic pregnancy? Am J Obstet Gynecol. 2010;202(6):630.e1-630.e5. doi:10.1016/j.ajog.2010.03.022
    Article · PubMed · Google Scholar
19. Silva C, Sammel MD, Zhou L, Gracia C, Hummel AC, Barnhart K. Human chorionic gonadotropin profile for women with ectopic pregnancy. Obstet Gynecol. 2006;107(3):605-610. doi:10.1097/01.aog.0000198635.25135.e7
    Article · PubMed · Google Scholar
20. Ferreira DF, Elito Júnior J, Araujo Júnior E, Stavale JN, Camano L, Moron AF. Trophoblastic infiltration in tubal pregnancy evaluated by immunohistochemistry and correlation with variation of beta-human chorionic gonadotropin. Patholog Res Int. 2014;2014:302634. doi:10.1155/2014/302634
    Article · PubMed · Google Scholar
21. Oktay K, Brzyski RG, Miller EB, Krugman D. Association of serum beta-hCG levels with myosalpingeal invasion and viable trophoblast mass in tubal pregnancy. Obstet Gynecol. 1994;84(5):803-806.
    PubMed · Google Scholar
22. Pulatoglu C, Dogan O, Basbug A, Kaya AE, Yildiz A, Temizkan O. Predictive factors of methotrexate treatment success in ectopic pregnancy: a single-center tertiary study. North Clin Istanb. 2018;5(3):227-231. doi:10.14744/nci.2017.04900
    Article · PubMed · Google Scholar
23. Dudley PS, Heard MJ, Sangi-Haghpeykar H, Carson SA, Buster JE. Characterizing ectopic pregnancies that rupture despite treatment with methotrexate. Fertil Steril. 2004;82(5):1374-1378. doi:10.1016/j.fertnstert.2004.03.066
    Article · PubMed · Google Scholar
24. Cohen A, Bar-On S, Cohen Y, et al. Ruptured ectopic pregnancies following methotrexate treatment: clinical course and predictors for improving patient counseling. Reprod Sci. 2022;29(4):1209-1214. doi:10.1007/s43032-022-00881-7
    Article · PubMed · Google Scholar
25. da Costa Soares R, Elito J Jr, Camano L. Increment in beta-hCG in the 48-hour period prior to treatment: a new variable predictive of therapeutic success in the treatment of ectopic pregnancy with methotrexate. Arch Gynecol Obstet. 2008;278(4):319-324. doi:10.1007/s00404-008-0589-4
    Article · PubMed · Google Scholar

Tables

Additional Information

Publisher’s Note
Bayrakol MP remains neutral with regard to jurisdictional and institutional claims.

Rights and Permissions