Abstract
Purpose: :
To compare capsulorhexis performance metrics between the Sensimmer phaco simulator and live surgery performed by PGY 4 residents.
Methods: :
Three performance metrics were developed: time to complete capsulorhexis; number of capsular grabs per completed rhexis; and circularity of capsulorhexis. Circularity index was ratio of average radius at 45 deg intervals to maximal radius in capsulorhexis trajectory. Twelve PGY 4 residents from 4 training programs were recruited. Performance metrics were extracted from 3 video recordings of non-complicated cataract surgery cases from each resident; and during the same time frame on the simulator from these simulator-naïve residents who were given brief simulator training prior to testing in 2 to 4 trials. Averages, standard deviations and correlations using a two-sample t-test were calculated.
Results: :
Average simulator completion time was 96.91 ± 44.23 sec and it was 94.42 ± 65.74 sec during surgery. Mean number of grabs on the simulator was 10.66 ± 4.81 and during surgery it was 10.31 ± 5.23. Circularity index of capsulorhexis on the simulator was 0.92 ± 0.04 and in surgery it was 0.88 ± 0.04 . Although average completion times and number of grabs on the simulator were comparable between the two groups, there was considerable variability between individual residents. Surgeons who showed the greatest variability on the simulator for these two metrics appeared to show increased variability during surgery while those residents that demonstrated less variability on the simulator showed reduced variability during actual surgery. For all surgeons, there was a high correlation between the circularity index on the simulator and in surgery (p=0.0002).
Conclusions: :
PGY4 residents who completed comparable numbers of surgical cases showed considerable variability during surgery in duration of capsulorhexis and in number of forceps grabs. The simulator was able to identify this variability. Average performance measures of a group may be misleading and may not reflect surgical performance of an individual resident. Rhexis circularity with simulator correlated best with surgical performance suggesting that this metric may be useful. Additional training tools and techniques were needed to try to reduce performance variability in residents- in- training.
Keywords: training/teaching cataract surgery