**3.1 Prevention of PEP**

One hundred two patients were enrolled in the present study. The prophylaxis and control groups consisted of 55 and 47 patients, respectively, with no intergroup differences in age, gestational age, or preoperative hCG levels. PEP were not noted in the prophylaxis group, but occurred in 8 patients (17.0%) in the control group (p<0.05; Table 1).

The 8 patients in the control group who developed PEP received a single systemic administration of MTX (50 mg/m2) between postoperative days 7 and 10, the period during which the diagnosis was made. One patient had a poor decline in serum hCG and required an additional administration of MTX (50 mg/m2) 7 days later. In addition, another patient developed lower abdominal pain and a hemoperitoneum 4 days after MTX administration, and underwent laparoscopic salpingectomy. The remaining six patients had a steady decline in serum hCG levels. Patients in both groups who did not develop PEP reached undetectable serum hCG levels, and thus completed the recommended follow-up by postoperative day 28. In contrast, patients who developed PEP required a mean follow-up of 51.7±17.2 days (p<0.05; Table 2).

No side effects attributable to MTX, such as dermatitis, alopecia, dyspepsia, and hepatic or bone marrow toxicity, were observed in the prophylaxis group.

## **3.2 Early detection of PEP**

In 42 of the 53 patients, no postoperative symptoms of PEP were noted, and the serum hCG levels dropped to pre-pregnancy levels; the PEP was located in 11 patients. Table 3 presents the clinical characteristics of both groups. No statistically significant differences existed between the PEP and control groups with respect to age, parity, gravidity, gestational age, specimen diameter, or preoperative serum hCG levels.


Persistent Ectopic Pregnancy After Laparoscopic

mg/m2) was given.

group than the control group.

Linear Salpingostomy for Tubal Pregnancy: Prevention and Early Detection 101

Figure 1 shows the postoperative hCG declines in the PEP group. In one patient, the postoperative serum hCG levels steadily increased post-operatively, and in the other nine patients, the levels of serum hCG decreased transiently, then increased. After period C, the hCG levels of all patients in the PEP group increased and did not decrease until the second intervention. In these patients, systemic administration of one additional dose of MTX (50

Figure 2 compares the variance in the hCG decline for each study period between the control and PEP groups. After period B, the hCG decline was significantly less in the PEP

Fig. 2. Comparison of the variance in the serum human chorionic gonadotropin (hCG) decline for each period between the control and persistent ectopic pregnancy (PEP) groups.

Figure 3 presents the 95% confidence interval (CI) of the hCG decline for 1 week after surgery in the control group, and the postoperative hCG decline in the PEP group. In the PEP group, the hCG decline after period C was outside the 95% CI of the control group. Analysis by ROC, sensitivity and specificity were calculated with optimal points in each period, and 14% of preoperative hCG valued in period C and D revealed that the specificity

Data are presented as median value and interquartile ranges (IQR) in each period.

and sensitivity of the test were equal to 100% (Figure 4).




Fig. 1. Postoperative declines in serum human chorionic gonadotropin (hCG) levels (percentages of preoperative hCG levels) during the first week after laparoscopic salpingotomy in the persistent ectopic pregnancy patients.

Table 3. Clinical characteristics of subjects in the study of postoperative declines in serum human chorionic gonadotropin (hCG) levels and persistent ectopic pregnancy (PEP).

Fig. 1. Postoperative declines in serum human chorionic gonadotropin (hCG) levels (percentages of preoperative hCG levels) during the first week after laparoscopic

salpingotomy in the persistent ectopic pregnancy patients.

Table 2. Duration of follow up.

Figure 1 shows the postoperative hCG declines in the PEP group. In one patient, the postoperative serum hCG levels steadily increased post-operatively, and in the other nine patients, the levels of serum hCG decreased transiently, then increased. After period C, the hCG levels of all patients in the PEP group increased and did not decrease until the second intervention. In these patients, systemic administration of one additional dose of MTX (50 mg/m2) was given.

Figure 2 compares the variance in the hCG decline for each study period between the control and PEP groups. After period B, the hCG decline was significantly less in the PEP group than the control group.

Fig. 2. Comparison of the variance in the serum human chorionic gonadotropin (hCG) decline for each period between the control and persistent ectopic pregnancy (PEP) groups. Data are presented as median value and interquartile ranges (IQR) in each period.

Figure 3 presents the 95% confidence interval (CI) of the hCG decline for 1 week after surgery in the control group, and the postoperative hCG decline in the PEP group. In the PEP group, the hCG decline after period C was outside the 95% CI of the control group. Analysis by ROC, sensitivity and specificity were calculated with optimal points in each period, and 14% of preoperative hCG valued in period C and D revealed that the specificity and sensitivity of the test were equal to 100% (Figure 4).

Persistent Ectopic Pregnancy After Laparoscopic

level than patients without PEP.

salpingostomy before PEP is ruled out.

in the PEP group (approximately 3-4 days postoperatively).

**4. Discussion** 

Linear Salpingostomy for Tubal Pregnancy: Prevention and Early Detection 103

As indicated by the results of the present study, prophylactic local administration of MTX into the tubal wall immediately after linear salpingostomy is extremely effective in preventing post-operative PEP. In addition, even when systemic MTX administration was effective, patients who developed PEP required a significantly longer follow-up of hCG

MTX has been shown to have no adverse effects on future pregnancies as long as pregnancy is avoided for a certain period of time,15 and may thus be proactively used as prophylaxis following salpingostomy. Graczykowski et al.7 reported that the incidence of PEP was reduced to 1.9% following a single systemic administration of prophylactic MTX (1 mg/kg) within 24 hours after salpingostomy. However, although generally mild, side effects related to systemic MTX administration have been reported in up to 24% of cases,8 including some cases of serious side effects.9 Therefore, the implementation of prophylactic systemic administration of MTX for all patients remains controversial. Importantly, local intratubal administration of MTX has been reported to enhance local anti-trophoblastic activity,16 in addition to reducing side effects,17,18 and may thus be a more effective and safer regimen for preventing PEP. This assumption is supported by the fact that no cases of PEP or side effects were observed in the MTX group in the present study, while PEP was observed in 1.9% of

Regarding the toxicity to the tube of local administration of MTX (50 mg), no effects were reported in a histologic study of intratubal injection of MTX (100 mg).19 Furthermore, subsequent fertility after local MTX injection was satisfactory.13,20 Therefore, local

Administration of prophylactic MTX to all patients remains a controversial issue. Prophylactic administration of MTX may be appropriate for patients at increased risk for developing PEP, such as patients with a short duration of amenorrhea, a small ectopic pregnancy (< 2 cm in size), and a preoperative hCG level ≥ 2500 IU/ml.21 Considering that no side effects were reported after local MTX administration in the present study, and in light of the risk of salpingectomy and the need for long follow-up in cases of PEP, it may be

This study also showed that age, parity, gravity, gestational age, specimen diameter, and preoperative serum hCG levels are not predictive of a PEP following a laparoscopic salpingostomy. Several attempts have been made to predict a PEP; however, no effective predictive protocols for PEP currently exist.22 Because our results were comparable and decreasing pattern of serum hCG has been reported to be helpful aid in avoiding further surgery,23 serum hCG levels must be closely monitored in all patients who have had a

In the current study, no difference existed in the decline in serum hCG postoperatively between the PEP and control groups during period A; however, after period B, the decline in serum hCG in patients with PEP was significantly less than the control group. This finding indicates that a subsequent increase in the serum hCG level occurs during period B

Previous studies have used the decline in serum hCG to detect a PEP24-26; however, all of the studies have used a single early post-operative hCG measurement. We indicated that the future course of serum hCG cannot be predicted reliably from a single early postoperative measurement. In our study, during period B, the hCG decline in patients with PEP began to be less than the 95% CI of the control group, and from period C, the decline in PEP group was

cases in a study involving single systemic administration of MTX.8

administration of MTX (50 mg) was thought to have no toxicity in the tube.

appropriate to consider prophylactic local administration for all patients.

Fig. 3. Postoperative declines in serum human chorionic gonadotropin (hCG) levels during the first week after laparoscopic salpingotomy in the successfully treated patients (control group, black line [95% confidence interval]) an PEP patients (Black boxes [individual hCG declines]). Between the black lines is presented 95% confidential interval of the control group. Black boxes present individual preoperative hCG values.

Fig. 4. Analysis by receiver operating characteristic curve correlating sensitivity of the test with the false-positive rate (1-specificity) for each postoperative period for the first week following laparoscopic salpingotomy. Sensitivity and specificity were calculated with optimal points in each period, and the excellent sensitivity and false positive rate (1 specificity) were plotted on this figure. A 14% of preoperative serum human chorionic gonadotropin (hCG) value in period C and D revealed that the specificity and sensitivity of the test were equal to 100%.
