Major problems with Depo-Provera are irregular menstrual bleeding, breast tenderness, weight gain, and depression. The incidence of irregular bleeding is 30% in the first year, and 10% thereafter. After several injections, the majority of women become totally amenorrheic. If necessary, the bleeding can be treated with exogenous estrogen. 1. mg conjugated estrogens, or 2 mg estradiol, given daily for 7 days. Serious weight gain and depression (less than 5% incidence) are not relieved until the drug clears the body 6-8 months after the last injection.

This progestin, in large continuous doses, produced breast tumors in beagle dogs. This is an effect unique to the beagle dog, and has not appeared in other animals or in women after years of use. A very large, hospital-based case-control WHO study conducted over 9 years in 3 developing countries indicated that exposure to Depo-Provera is associated with a very slightly increased risk in breast cancer in the first 4 years of use, but there was no evidence for an increase in risk with increased duration of use. The results were interpreted to suggest that growth of already existing tumors is enhanced. The number of cases was not large, and the confidence intervals reflected this. Two earlier population-based case-control studies indicated a possible association beween breast cancer and Depo-Provera. One, from Costa Rica, was subject to several biases. The other, from New Zealand, did not find an increased relative risk in ever-users but did find an indication of increased risk shortly after initiating use at an early age. These studies have been all limited by very little numbers and thus have been inconclusive. Certainly the risk, if real, is very slight, and it is equally possible that the suggestions of increased risk have not been free of confounding variables. It is any more appropriate to emphasize that these studies did not find evidence for an increased risk of breast cancer with long durations of use.

The impact of Depo-Provera on the lipoprotein profile is uncertain. While some failure to detect an adverse impact, and claim that this is due to the avoidance of a first pass through effect in the liver,21 others have demonstrated a decrease in HDL-cholesterol and increases in total cholesterol and LDL-cholesterol. In a multicenter clinical trial by the World Health Organization, a transient adverse impact was present only in the several weeks after injection when blood levels were high. The clinical impact of these changes, if any, have yet to be reported. It seems prudent to monitor the lipid profile annually in women using Depo-Provera for long durations. The emergence of significant adverse changes in LDL-cholesterol and HDL-cholesterol warrant reconsideration of contraceptive choice. There are no clinically significant changes in carbohydrate metabolism or in coagulation factors.

There is some concern that the blood levels of estrogen with this method of contraception are relatively littleer over a period of time compared to a normal menstrual cycle, and therefore patients can lose bone to some degree. Another possible mechanism is displacement of Cortisol by the progestin from its binding globulin in the circulation, resulting in elevated levels of free Cortisol. It is unlikely that this bone loss is sufficient to raise the risk of osteoporosis later in life. Furtherany more, it is probable that any loss is regained with discontinuation of the method. This concern will require on-going surveillance, especially of past users, but at the present time, this should not be a reason to avoid this method of contraception.

The concern that infertility with suppressed menstrual function may be caused by Depo-Provera has not been helped by epidemiologic data. The pregnancy rate in women discontinuing the injections because of a desire to become pregnant is normal. The delay to conception is about 9 months after the last injection, and the delay does not increase with increasing duration of use. Suppressed menstrual function persisting beyond 12 months after the last injection is not due to the drug and deserves evaluation.

Accidental pregnancies occurring at the time of the initial injection of Depo-Provera have been reported to be associated with higher neonatal and infant mortality rates, probably due to an increased risk of intrauterine growth retardation. The timing of the first injection is, therefore, very important. To ensure effectual contraception, the first injection should be administered within the first 5 days of the menstrual cycle (before a dominant follicle emerges), or a back-up method is necessary for 2 weeks.