Retrospective analysis of laparoscopic salpingostomy and conservative expectant management of tubal ectopic pregnancy

Abstract

Purpose

To identify predictive factors for successful expectant management of ectopic pregnancy and to evaluate the prognosis for fertility after expectant management and laparoscopic salpingostomy.

Methods

Forty‐six cases of expectant management and eighty cases of laparoscopic salpingostomy for tubal ectopic pregnancy were retrospectively analyzed. Subjects were classified in three groups: those who underwent laparoscopic salpingostomy, those treated by expectant management only, and those treated by expectant management but requiring additional treatment.

Results

The rates of tubal patency, intrauterine pregnancy and repeated ectopic pregnancy in the laparoscopic salpingostomy group were 75, 40, and 16%. The rates in the expectant management group were not significantly different: 72, 42 and 15%. Finally, the rates in the extra treatment group were 75, 39 and 15%. Success rate of expectant management was 54%. In 93% of cases expectant management was successfully completed when the initial levels of urinal hCG were less than 3000 mIU/ml and the levels of hCG 48 h later were less than 80% of the initial levels. However, expectant management alone was insufficient and required extra treatment in 90% of cases when the initial levels of hCG were 3000 mIU/ml and above or when the levels of hCG level 48 h later was 80% of initial levels and above.

Conclusions

Expectant management in combination with salpingostomy is not only minimally invasive but also a useful way to preserve fertility. Initial urine hCG levels and their variation over time can help predict whether expectant management will succeed.

Introduction

Laparoscopic treatment and methotrexate administration are often introduced for ectopic tubal pregnancy because earlier diagnosis is now possible using highly sensitive hCG assays and transvaginal sonographs. When circulating hCG levels are stable, even conservative management can be chosen under certain conditions. So far, many cases have been reported in which conservative expectant therapy has succeeded. Thus, some of our cases with minor symptoms opted for expectant management. However, some cases enrolled in expectant management needed further surgical treatment or methotrexate administration. It is still unknown what the predictive factors are for success of expectant management. We have retrospectively analyzed clinical features in cases of expectant management and laparoscopic salpingostomy for tubal ectopic pregnancy in an attempt to clarify this situation.

Materials and methods

Forty‐six cases of expectant management and eighty cases of laparoscopic salpingostomy for tubal ectopic pregnancy were retrospectively analyzed. All cases were diagnosed as tubal ectopic pregnancy between June 1995 and November 2005 in our institute, Yokohama City University Medical Center and Yokohama City University Hospital. Our protocol for tubal ectopic pregnancy is shown in Fig. 1. Cases were diagnosed as tubal pregnancy by transvaginal sonograph with minimal Douglas fluid as no fetal heart beat, and minimal abdominal pain. The diagnosis of ectopic pregnancy is sometime difficult, when hCG level is low. The enrolled cases with lower hCG level were confirmed for formation of tubal hematoma by transvaginal sonography or absence of villi in a uterine cavity by surgical curettage. Patients granting consent (with information about the risk of sudden hemorrhage) were assigned to expectant management. On the other hand, symptomatic cases desiring a child with no history of ectopic pregnancy, an unruptured focus, less than 5 cm focus diameter, absence of fetal heart beat and less than 10000 mIU/ml urine hCG levels, were enrolled in the laparoscopic linear salpingostomy arm of the study. Of the cases enrolled into the laparoscopic linear salpingostomy thirty‐five had hCG level less than 3000 mIU/ml. The characteristics of enrolled cases are shown in Table 1.

Figure 1.

Flow chart for treatment of ectopic pregnancy

Table 1.

Clinical characteristics of enrolled cases

	Laparoscopic salpingostomy (n = 80)	Expectant management (n = 46)	p value
Age (years)	29.2 ± 5.4	30.3 ± 5.1	0.58
Gravida	1.3 ± 1.6	1.6 ± 1.5	0.36
Parity	0.3 ± 0.7	0.5 ± 0.6	0.11
Gestational weeks	6.3 ± 1.4	6.5 ± 1.3	0.22
Initial urine hCG (mIU/ml)	3179 ± 1546	1744 ± 1146	0.001

Values are mean ± standard deviation

Expectant management cases were examined by transvaginal sonograph and the level of urine hCG was measured every 48 h. Urine hCG level was determined using the latex agglutination assay (Mitsubishi Kagaku, Tokyo) and correlated with urine creatinine level. The cases in which hCG levels declined to less than 15 mIU/ml were classified as cured. In some cases of expectant treatment, laparoscopic surgical treatment was performed subsequently, when the level of hCG was not in decline or when peritoneal hemorrhage was visible by sonograph. In the cases of linear salpingostomy, tubectomy or methotrexate administration, the procedure was performed when the hCG level had maintained at higher than 10% of the initial level, even 7 days after the linear salpingostomy.

In all cases, hysterosalpingography was performed 3 months after the initial treatment to assess patency of uterine tubes.

Statistical significance of variables were analyzed with the χ² test or Student's t test.

Results

Seventy‐four cases (93%) of laparoscopic salpingostomy were uneventful. Three cases (3.5%) needed laparoscopic tubectomy and another three (3.5%) cases were treated by methotrexate administration for regulation of persistent ectopic pregnancy after salpingostomy. On the other hand, 26 cases (57%) of expectant management were completed without any major complications. However, fourteen cases (30%) needed laparoscopic salpingostomy or laparoscopic tubectomy because of peritoneal bleeding or prolonged high hCG levels. Systemic methotrexate administration was performed in 6 cases (13%) for persistently high levels of hCG. Post‐surgical hysterosalpingography showed that in 75% of the cases undergoing linear salpingostomy, the treated side of the tubes maintained their patency. Seventy‐five percent of successfully completed expectant management cases and 72% of cases requiring extra treatment also maintained tubal patency (Table 2).

Table 2.

Fertility prognosis after treatment

	Tubal patency (HSG) (%)	Intrauterine pregnancy (%)	Repetitive ectopic pregnancy (%)
Laparoscopic salpingostomy (n = 80)	75	42	16
Expectant management (successful cases) (n = 26)	72	40	15
Expectant management (required extra treatment) (n = 20)	75	39	15

The pregnancy ratios within one year of the treatment were 42, 40 and 39% in the cases of salpingostomy, expectant management and expectant management with extra treatment, respectively. The ratios of repetitive ectopic pregnancy were 16, 15 and 15% in the case of salpingostomy, expectant management and expectant management with extra treatments, respectively (Table 2). Table 3 shows the clinical characteristics of the cases of expectant management. The mean initial level of urine hCG in the successfully completed cases were significantly lower than the initial hCG levels in the cases requiring extra treatment.

Table 3.

Clinical characteristics of expectant management

	Successful cases (n = 26)	Required extra treatment (n = 20)	p value
Age (years)	29.5 ± 5.5	28.9 ± 5.3	0.82
Gravida	0.76 ± 1.3	0.57 ± 1.1	0.63
Parity	0.04 ± 0.2	0.10 ± 0.3	0.44
Gestational weeks	6.36 ± 1.3	6.44 ± 1.6	0.92
Initial urine hCG (mIU/ml)	876 ± 647	1390 ± 1522	0.03

Values are mean ± standard deviation

The initial level of urinal hCG was less than 3000 mIU/ml in all successful cases. The mean level of urine hCG at 48 h falls to 56% of its initial level in the successful group. On the other hand, urine hCG levels were elevated in almost 60% of the cases in need of extra treatment. In 93% (24 cases) of cases, expectant management was successfully completed when the initial level of urine hCG was less than 3000 mIU/ml and the level of hCG 48 h later was less than 80% of its initial level. On the other hand, 90% of expectant management ‘failed’, requiring extra treatment (e.g., additional laparoscopic tubectomy or methotrexate administration) when the initial level of hCG was 3000 mIU/ml and above or when the hCG level 48 h later was 80% of initial levels and above (Table 4).

Table 4.

Relationship between success and hCG levels in expectant management

	Successful cases	Required extra treatment
A	24	2
B	2	18

Values are number of cases

A, initial urine hCG level <3000 mIU/ml and decreases >20% (after 48 h); B, others

Discussion

Langer [1] and colleagues reported that the frequency of repetitive ectopic pregnancy after laparoscopic salpingostomy was 12.7% and that 47% of repetitive ectopic pregnancy occurred on the intact side of the tube. It also suggested that tubectomy was not helpful in the prevention of repetitive ectopic pregnancy. Pouly [2] investigated the fertility of cases that preserved tubes after surgical treatment for ectopic pregnancy. The ratios of intra‐uterine pregnancy and ectopic pregnancy after salpingostomy were 67 and 12%, respectively. Furthermore the ratio of intrauterine pregnancy and the ratio of ectopic pregnancy were 40 and 30%, respectively, even in the cases where lateral tubes had been removed. Thus, we surmise that salpingostomy may be effective for maintaining fertility.

Expectant management for ectopic pregnancy was first reported by Lund [3]. It is well known that some cases of ectopic pregnancy could be managed conservatively without severe complications. However, indication for expectant management is not yet standardized. Some investigators suggested that the initial serum level of hCG (1000 mIU/ml [4, 5, 6] to 2000 [7] mIU/ml) is a predictive factor for the success of expectant management. Furthermore, Korhonen [8] reported that longitudinal changes in hCG level might be predictive of successful expectant management.

It has been reported that 50–70% [4, 5, 6, 7, 8] of expectant management cases are successful. In our study, 54% of enrolled expectant management cases were completed uneventfully, consistent with previous reports. In cases of expectant management, the mean initial level of urinal hCG was 876 mIU/ml. In all cases of successful expectant management, the initial urine hCG was less than 3000 mIU/ml and the level of urinal hCG 48 h later decreased from initial levels. On the other hand, in almost all of the failed cases, hCG levels did not decrease. Even when the levels of hCG 48 h later had decreased up to 80% of its initial levels, expectant management failed in three cases. We suggest that both the initial level of hCG and its fluctuation may be a predictive factor in successful expectant management or the decision to shift from expectant management to additional treatment. Serum progesterone levels [9, 10] and the dynamics of arterial blood flow [11] have been reported as other predictive factors for successful expectant management. However, these factors are still in need of further investigation.

It has been reported [12] that the ratio of tubal patency of the affected side after expectant management, was 93%. In the same report the ratio of normal pregnancy and the ratio of repetitive tubal pregnancy within one year of the expectant management were 83 and 4%, respectively. Other investigators have also shown that the ratio of normal pregnancy and the ratio of repetitive tubal pregnancy within a year after the expectant management were 63 and 12%, respectively. These reports are consistent with our results.

Conclusion

Expectant management is not only a minimally invasive treatment, but also a useful way to preserve fertility with similar rates of success to other treatments, e.g., methotrexate therapy and salpingostomy. Expectant management must be one of the choices of treatment for tubal pregnancy when the initial urinary hCG level is less than 3000 mIU/ml and decreases to less than 80% of its initial level within 48 h.

Acknowledgments

We greatly thank Mario Ikeda, MD, PhD and Yoshihito Kondoh, MD, PhD for their kind support. No financial interest and no financial support were received.