Telesurgery with the Pakyrcm System

The PAKY-RCM arm has been successfully used as the first step in transcontinental PCNL between two countries in a few patients. On June 17,1998, the first remote telerobotic percutaneous renal access procedure was performed between the Johns Hopkins Hospital, Baltimore, Maryland, USA, and Tor Vergata University, Rome, Italy, a distance of some 11,000 miles [23]. Remote control of the robot was accomplished over a plain old telephone system line. Video connections were established using three ISDN lines on the Italian side switched to a T1 line in the United States. The telesurgical robot was successful in terms of obtaining percutaneous access within 20min, with two attempts to obtain entry into the collecting system.

In 2003 the group from Baltimore made a connection with a team in Sao Paulo, Brazil [25]. They described a laparoscopic bilateral varicocelectomy and a percutaneous renal access for PCNL. The technical setup consisted of a 650-MHz personal computer fitted with a Z360 video CODEC (coder/decoder) and a Z208 communication board (Zydacron, Manchester, NH, USA). This formed the core of the telesurgical station. In the PCNL patient, access to the urinary tract was achieved with the first needle pass, and percutaneous nephrolithotomy was uneventful. Blood loss was minimal, and the patient was discharged home on the second postoperative day.

Both of these initial clinical telerobotic procedures demonstrated the feasibility and safety of remote robotic needle access to the kidney during percutaneous procedures. Despite these successes, there were little quantitative data to scientifically support telerobotic PCNL in terms of speed and accuracy until a series of experiments between Johns Hopkins in Baltimore and Guy's Hospital in London [26], in 2002. In the first of these, half the needle insertions (152) were performed by a robotic arm (Fig. 4) and the other half by urological surgeons. The order was decided by the toss of a coin, except for a subgroup of 30 transatlantic robotic procedures. These robotic attempts were entirely controlled by a team at Johns Hopkins via four ISDN lines for video, sound, and robot data. The technical specifications were almost identical to the previous clinical case reports, as outlined above. A successful needle insertion was confirmed by passage of either a guidewire or contrast into the collecting system of the kidney model. For the robotic procedures, the operators viewed monitors showing both the robotic arm and a fluoroscopy image of the model in real time.

Fig. 4. PAKY-RCM during the transatlantic trial at Guy's Hospital, London

Fig. 4. PAKY-RCM during the transatlantic trial at Guy's Hospital, London

All needle attempts were successfully completed within three passes, with an interquartile range of 25 to 52 s (median, 35 s) for the human attempts compared with an interquartile range of 41 to 80s (median, 56s) for the robotic attempts. The robot was slower than the human operators to complete the insertions (p < 0.001, Mann-Whitney U test), but it was more accurate than the human operators because it made fewer attempts (the rate of success on the first attempt was 88% for the robot vs. 79% for the humans; p = 0.046, chi-squared test). All surgeons required fewer needle passes when using the robotic arm. The median time taken for transatlantic robotic needle insertion (59 s) was comparable to the median time taken for local robotic needle insertion (56 s), with no difference in accuracy.

In a second crossover trial [27], half of the needle insertions (30) were performed by a robotic arm in Guy's Hospital in London controlled by a team at Johns Hopkins in Baltimore via four ISDN lines; the other half of the needle insertions were controlled by the same robotic arm in the reverse direction. Again, all needle attempts were successful within two passes, with a median of 63 s for the Baltimore-to-London attempts compared with a median of 57 s for the London-to-Baltimore attempts (p = 0.266). There was no difference in accuracy between the trials controlled in different directions: the rate of first-pass accuracy was 84% for the Baltimore-to-London attempts, compared with 97% accuracy for the London-to-Baltimore attempts (p = 0.103). In comparison with the locally controlled robotic needle insertions, there was again no difference in time (median, 62s) or accuracy (91% rate of first-pass success).

From these trials one can conclude not only that telerobotic PCNL is feasible, but also that the robot is more accurate than the human hand, since it is significantly more successful on the first attempt. In addition, the remote robotic procedures compared favorably with local robotic and human procedures. The advantage of increased accuracy is maintained in both directions, and thus remote robot-assisted PCNL may have significant advantages in terms of accuracy and hence potential patient safety.

Was this article helpful?

0 0

Post a comment