I've had six miscarriages and one pregnancy to term. All my labs are normal, including antibodies. In your column on

Studies that show something works are likely to get published. Those that show something doesn't work are rarely published. I suggest you ask your doctor for any well-designed studies showing that progesterone has a beneficial effect. Those studies do not exist.

First, let's explore the rationale for using progesterone. After ovulation, progesterone helps prepare the uterine lining for implantation. When you conceive, the hCG hormone released by the developing placenta stimulates the ovaries to make progesterone until about six to eight weeks of pregnancy, when the placenta takes over the role of producing progesterone. Studies that look at miscarriage have shown that the level of hCG, which is responsible for progesterone production, always falls before the level of progesterone does. That indicates that a dropping progesterone level is a result of the failing pregnancy, not the cause. In fact, I have had patients who have successfully carried pregnancies with progesterone levels as low as 2ng/ml due to an enzyme block. Their very low level did not cause them to miscarry. Another issue is that giving hCG can sometimes stimulate progesterone levels. Some studies have shown a benefit giving hCG; others have not. Again, this suggests that a low level of hCG from the placenta is a sign of a failing pregnancy, not the cause.

We can see if the progesterone is preparing the uterine lining properly by taking an endometrial biopsy -- a snip of tissue from the uterine lining -- about 12-13 days from the LH surge. Still, up to 20 percent of biopsies taken from women who have no miscarriage problems are abnormal. So when I find abnormal biopsy results, I don't know if you are "normal" and just having a bad month or if you truly have a luteal phase defect (low progesterone effect on the uterine lining). To know this, I need to perform two biopsies. Studies suppressing FSH level early in the menstrual cycle find that luteal phase defect is most often caused by poor stimulation of the ovary before ovulation. This suggests that the proper treatment for luteal phase defect is ovulation induction, not progesterone supplementation.

I am frustrated by how many doctors and women just throw a bit of progesterone or baby aspirin at recurrent pregnancy loss, hoping all will be okay. That approach can often work. Statistically, placebos will be effective up to 50 percent of the time. That may explain your one success. While I don't believe "a bit of progesterone" makes any difference, progesterone may in fact turn out to be of benefit for selected women with recurrent pregnancy loss. In very high doses (400mg-600mg/day), progesterone may suppress natural killer cells, which can cause recurrent miscarriage.

You say all your labs are normal, including antibodies. Which labs? Do you have polycystic ovarian syndrome? If so, did they check insulin and glucose levels? If you are over 35, did they perform a clomiphene challenge test? Did they do a saline ultrasound? Did you have an endometrial biopsy -- or two? Did they measure a day 10 LH level, TSH or prolactin? How did they evaluate the immune system? Did they look for antibodies to DNA, histones, antiphosphoserine, ethanolamine, inositol and cardiolipin as well as doing a lupus anticoagulant test? Did they do a reproductive immunophenotype to measure the level of NK cells and carry out such tests as LAD, NK assay or embryotoxic factor? Did they get chromosomes on a prior loss? Did they send placental block tissue (available if you had a D&C with a miscarriage) to determine possible immune causes for miscarriage?

If the answers to any of these are no, or you don't know, you may be receiving care that banks on the 50 percent success rate using placebo and prayer.