Department of Obstetrics and Gynaecology, University of Liverpool, Liverpool L69 3BX, UK
Dear Sir,
Our study was a retrospective consecutive case series in which we analysed routinely collected data. We acknowledge methodological shortcomings of such an approach and agree with histopathologists that diagnosis, definitions and inter-/intra-observer variance of placental histology may be an important source of bias. We hoped that `routine' placental histology, often used to guide our clinical decision making, would be robust enough to find some characteristic placental changes in women with thrombophilia. We were wrong. Nevertheless our conclusions are by no means `definitive' and, like Khong et al. and Sebire, we are also looking forward to the large prospective studies with standardized protocols, internationally agreed definitions of thrombotic placental lesion and high internal and external consistency to confirm or refute our findings.
For the record, our study did not include diabetic women or women on folate, aspirin or heparin therapy. There were 19 smokers in our study group (24%), 11 of whom had had abnormal thrombophilia screen.
Notes
1 To whom correspondence should be addressed. E-mail: zarko{at}liverpool.ac.uk