Both the progestogen used and the dose of ethinylestradiol affect the venous thrombosis risk associated with combined oral contraceptive pills (COCPs), a systematic review and network meta-analysis has found.

The following databases were used: PubMed, Embase, Web of Science, Cochrane, CINAHL, Academic Search Premier and Science Direct up to April 2013. The search strategy included observational studies that covered healthy women using COCPs with a primary outcome of fatal or non-fatal first event of venous thrombosis especially deep vein thrombosis or pulmonary embolism. Publications were included with a minimum of ten venous thrombosis events in total.

Ten frequently prescribed COCPs were included in the network meta-analysis, see table 1.

The progestogens were classified into second generation (20LNG, 30LNG, 50LNG), third generation (20GSD, 30GSD, 20DSG, 30DSG and 35NRG) or unclassified (35CPA and 30DRSP).

A total of 3,110 publications were retrieved of which only 25 publications reporting on 26 studies were included. Incidence of venous thrombosis in non-users from two cohorts was 1.9 and 3.7 per 10,000 woman-years which is consistent with previously reported rates of 1-6 per 10,000 woman-years.

COCP use increased the venous thrombosis risk compared with non-use, relative risk (RR) 3.5, (95% CI:2.9 to 4.3). All preparations were associated with more than a two-fold increased risk of venous thrombosis compared with non-use.

The highest risk was seen with 50LNG (no longer available in the UK) and lowest in 20LNG and 20GSD users. The relative risk of venous thrombosis using COCPs containing 30/35 μg ethinylestradiol (EE) and GSD, DSG, CPA or DRSP were similar and were 50-80% higher than for LNG-containing COCPs.

A dose-related effect was noted for GSD, DSG and LNG; with higher doses associated with increased venous thrombosis risk.

This systematic review offers a comprehensive overview of the risk of venous thrombosis in women using common COCPs available in the UK. Past meta-analyses have used a 30 μg EE/150 μg LNG as a comparator with different pills which limits the interpretation. The network meta-analysis used in this study allowed a more direct comparison to be made between different COCPs.

Although this meta-analysis has not shed new light on venous thrombosis risks in different generation COCPs, it offers a way to make direct comparison with many COCPs available in the UK. One surprise for me is that although GSD is a third generation progestogen, the 20GSD pill appears to have low venous thrombosis risk similar to that of 30LNG.

The findings from this study should inform current good practice about choosing the best COCP.

It is important to take a good history (especially to assess risk of cardiovascular disease and venous thrombosis),

refer to the UK Medical Eligibility Criteria for risks and benefits, consider the non-contraceptive benefits and to ensure that the woman can make an informed choice.

Dr Richard Ma