Despite thousands of hand assisted laparoscopic donor nephrectomies (HALDNs) having been performed a paucity of data still exists in establishing how many procedures are required in order to safely ascend the learning curve. The production of such parameters is essential to the development of appraisal, safety and training programmes in this expanding surgical programme within transplantation. Recent data from other equivalent surgical procedures suggests both intraoperative and postoperative outcome data parameters can be used to establish the learning curve.

The caseload of two surgeons who had individually performed over 180 HALDNS was interrogated. Using cumulative sum analysis (CUSUM), operating time, hospital stay, the occurrence of major and minor complications and the need for readmission or reoperation were assessed. The learning curve was analysed using graphical representations to detect an inflexion point which would represent a stability of process. The number of procedures required to arrive at this point was assumed to represent successful ascent of the learning curve. Statistical analysis using the Pr > zL statistic was also used to quantify whether such a stability of process had been achieved.

Surgeons 1 and 2 performed 189 and 183 cases over an 8 year period. All were intraperitoneal HALDNs using a standardised technique. Patient demographics between the 2 surgical caseloads were similar (% female: 51 v 52%, mean age 44.1 v 44.7 yrs, mean BMI 26.5 v 27.2, % left sided cases 80 v 85%). CUSUM analysis revealed no discernible inflexion points for hospital stay (zL = 0.3 p=0.07), occurrence of Clavien 2 and above complications (zL =0.84, p=0.337), readmission (zL=0.696 p=0.243) or reoperation (zL= -0.366 p=0.643). Operating time however demonstrated a visible stability of process initially at case 25 but this was more sustained by case 40 to 45 for both surgeons.

A learning curve can be reproducibly established for HALDN using operating time as a surrogate marker. True ascent of the learning curve may mean the performance of up to 50 procedures rather than 20 – 25. In our series the occurrence of initial stability in operating times at case 25 was likely biased by careful patient selection. Hence the stability achieved by case 45 is likely more representative of a normal surgical HALDN casemix. We anticipate this data will better inform the development of an efficient and effective surgical training programme for HALDN.

CITATION INFORMATION: Ahmed Z, Tamburrini R, Uwechue R, Chandak P, Calder F, Kessaris N, Mamode N. Establishing a Learning Curve for Laparoscopic Living Donor Nephrectomy. Am J Transplant. 2016;16 (suppl 3).

« Back to 2016 American Transplant Congress