What can we learn from large data sets? An analysis of 19,000 retropubic tapes

Retropubic tapes are successful for treating stress urinary incontinence (SUI), but there is controversy around risk profiles. The British Society of Urogynaecology (BSUG) database allows analysis of surgery for patient safety, surveillance and benchmarking. Objectives of this study were to establish success and complication rates in routine practice, determine complication rates for trainees and consultants, explore reasons for outliers and assess perforation as a surrogacy of quality.

Approval was obtained from BSUG to use data on retropubic tapes. Data was anonymised, and patients gave prior consent. Analysis was done using the χ² test, and a funnel plot of bladder perforation rate was calculated.

There were 18,763 procedures recorded: 14,156 were performed by consultants, 64 by associate specialists (64), 1140 by subspecialty trainees, 2549 by registrars, 201 staff grades and 377 other. We found a 3.5 % bladder perforation rate, which was statistically higher for trainees than consultants (p < 0.05). The rate of other "standard" complications were low: 95.8 % of patients felt better on the Patient Global Impression of Improvement of Incontinence (PGI) scale. There was a significant difference (p < 0.05) in PGI and SUI outcome between patients who did and did not experience perforation.

Success rates with retropubic tapes are high, with low complication rates. Bladder perforation in "real", not "trial" data was 3.5 %, which is lower than reported by the Cochrane review (4.5 %). Trainees have a higher perforation rate (p < 0.05) because of learning curves. Outliers are easily identified, and reasons for this should be explored, including proportion of trainees doing the surgery. This analysis confirms that bladder perforation is a valid surrogate for quality with a small but measurable difference. We have a responsibility to analyse data to improve patient care and encourage colleagues to support the International Urogynaecology Association (IUGA) database initiative.