Re: use of serum progesterone measurements to reduce need for follow‐up in women with pregnancies of unknown location

Abstract

We read with great interest the paper by Day et al.1. We congratulate the authors on reducing the need for unnecessary follow-up. Specifically, Day et al. investigated the feasibility of a single-visit strategy in pregnancies of unknown location (PULs) which are at low risk of intervention. They concluded that a progesterone cut-off of < 10 nmol/L as the sole diagnostic test for spontaneously resolving PULs would eliminate the need for routine follow-up in nearly 50% of women. However, 12% of ectopic pregnancies which would eventually require intervention would be discharged. Any protocol which reduces the need for follow-up in PULs must be balanced against any possible harm, i.e. discharging women with potential ectopic pregnancies after a single visit without follow up. In 2003, we published on the use of the same single-visit protocol that is advocated by Day et al. in the management PULs2. In that study, we concluded that although 74.7% of low-risk PULs were correctly being discharged without further follow-up, this approach to the management of PULs was unsafe, with 49% of ectopic pregnancies being discharged. We believe that a single-visit strategy should not be used as an alternative to the current two-visit strategy which incorporates the human chorionic gonadotropin (hCG) 48 h/0 h ratio3-9. Even if 15% of ectopic pregnancies can be managed safely expectantly, i.e. the self-limiting forms of ectopic pregnancy, the rate of discharged ectopic pregnancies is still disproportionately high. In Day et al.'s study, approximately half of all interventions were surgical evacuations of miscarriages, the other half including either surgery or medical management of ectopic pregnancies. A significant proportion of PULs which developed into non-viable intrauterine pregnancies seemed to consist of incomplete miscarriages. In our experience, if a woman is classified on the primary ultrasound examination as having a PUL (i.e. no evidence of an intra- or extrauterine pregnancy and an absence of retained products in a woman with a positive urinary pregnancy test), it is unlikely that on subsequent scans she will have an incomplete miscarriage confirmed (i.e. with evidence of retained tissue on ultrasound). In Day et al.'s study, incomplete miscarriages at the first scan seem to have been pooled together with PULs. Incomplete miscarriages should not be managed as PULs with determination of serum progesterone/serial hCG levels. In summary, we agree with the philosophy of reducing intervention in the PUL population. This, however, must not be at the expense of maternal safety. In 2003, we published data2 on the use of the protocol advocated by Day et al. and in 2005, we published data on the use of a new single-visit strategy to manage women with a PUL9. In both of these studies, in which serum progesterone at presentation was used in combination with serum hCG, we concluded that a single-visit approach to the management of PUL was not safe practice. Based upon our previously published results, we believe this still to be the case. The results of the study of Day et al. should be interpreted with caution. T. Bignardi*, G. Condous*, * Early Pregnancy, Acute Gynaecology and Advanced Endosurgery Unit, Nepean Clinical School, University of Sydney, Sydney, Australia