Single centre versus multi-centre pooled morbidity data in PCNL and the implications for informed consent

Abstract

Background: National and international outcome data following PCNL have been available for many years, but multi-centre data may not reflect the outcome from an individual surgeon or hospital.Methods: A combination of retrospective and prospective single centre data was collated from 2000–2016 and are compared to large single and multi-centre series.Results: Data were available on 801 unique cases performed between 2000 and 2016, mean age = 55.2 (SD = 14.8) (range = 17–93). The mean change in haemoglobin after PCNL was 1.65 g/dL ± 0.05, n = 630. Twenty-seven patients required a blood transfusion (3.37%). In 470 cases, data on pre-operative urine culture was available. One hundred and nineteen (25%) demonstrated evidence of bacteriuria pre-operatively. The most common isolated species were E. Coli and Proteus Mirabilis. Pre-operative urine infection was associated with a greater drop in haemoglobin following surgery, but this difference was not found to be statistically significant. Changes in serum creatinine and eGFR rise following surgery were calculated. The mean rise was found to be 15.21 µmol/L (SE = 2.08, n = 208). The mean drop in eGFR was estimated to be 7.35 ml/min/1.73 m2 (± 0.895, n = 205). Eight cases of 801 (1%) required admission to higher level care. There was one small bowel puncture and one pleural perforation recorded. Sub-selective embolization due to bleeding occurred in six cases (0 .75%) and there were no peri-operative deaths in this series. Published data comparing single centres with > 500 cases are presented.Conclusion: To facilitate transparent consent, single-centre rather than pooled outcome data should be utilized.