Sixty-nine cases of PUL with a positive pregnancy test and haemoperitoneum but unknown location of the gestational sac at first sight on exploratory surgery were analysed at the Department of Obstetrics and Gynaecology, Shanghai Jiaotong University. The level of hCG in the haemoperitoneum and venous serum were measured, and the ratio of hCG in haemoperitoneum and venous serum (Rp/v-hCG) was calculated. Rp/v-hCG < 1.0 was taken to suggest an intrauterine gestational sac, and Rp/v-hCG > 1.0 was taken to suggest an abdominal gestational sac. The sensitivity and specificity of Rp/v-hCG for predicting the location of the gestational sac were evaluated prospectively.
Among 69 cases of PUL, 17 cases (17/69) were ultimately diagnosed as abdominal gestational sacs before 9 weeks of gestation, and 52 cases (52/69) were ultimately diagnosed as intrauterine gestational sacs. The diagnostic sensitivity and specificity of Rp/v-hCG at the time of exploratory surgery for predicting the location of the gestational sac were 94.1 % and 100 % , respectively (kappa = 0.96; P = 0.039). The rate of missed diagnosis was 5.9 % . The location of the gestational sac was determined during the initial exploratory procedure for 15 cases (15/17) with an abdominal gestational sac (1 case of splenic pregnancy was diagnosed during secondary surgery) and 37 cases (37/52) with an intrauterine gestational sac. With the exception of gestational sacs located in the pouch of Douglas (52.9 % , 9/17), the gestational sacs (47.1 % , 8/17) located in the other places were difficult to find.
Rp/v-hCG should be considered when exploratory surgery reveals no visible gestational sacs at first sight. If Rp/v-hCG is >1.0, more careful pelvic or abdominal exploration is required, rather than dilation and curettage, to locate abdominal gestational sacs.