MAUDE Adverse Event Report: BLUE BELT TECHNOLOGIES REAL INTELLIGENCE CORI; ORTHOPEDIC STEREOTAXIC INSTRUMENT

Model Number ROB10024

Device Problem Application Program Problem: Parameter Calculation Error (1449)

Patient Problem No Clinical Signs, Symptoms or Conditions (4582)

Event Date 06/01/2021

Event Type  Injury  

Event Description

It was reported that during a cori assisted ukr surgery the real intelligence robotic drill had a disconnect error.They proceeded to quit the case, unplug, re-plug and recalibrate (case-(b)(4)), but the bone model appeared to be distorted and the surgeon decided to proceed, but the model showed some areas of red resection.The system was showing accurate information as the surgeon had cut extra bone (case-(b)(4)).They ensured the trackers did not move and adjusted the plan slightly.This allowed them to correct the resection and complete a successful case.The procedure was completed, with a delay of fewer than 30 minutes with the same devices.No patient injuries and no other complications were reported.

Manufacturer Narrative

Section h10: the real intelligence cori used for treatment was not returned to the designated complaint unit for evaluation, therefore, visual and functional inspections could not be performed.The product was not returned, however, the software files were downloaded from the device and provided for investigation.The log files do not indicate any system or software issue related to the reported complaint.The screenshots were reviewed with clinical account representatives.The screenshot review confirmed the reported areas of red resection on the ¿bone removal-femur¿ screen.The red found on the anterior edge of the condyle is likely due to over-modeling of the bone.Less free collection is required when working with a smaller area of the bone, which is the case for ukr surgeries.Ukr requires point collection of the articular surface and the landmark points bounding the condyle.Therefore, with this smaller area of the bone being collected, a starter model is used to generate the bone removal model that initially appears in bone refinement, and subsequently in bone removal if the bone model had not been refined.This starter model is larger than the free collection area collected and the implant.This ensures that all areas where the bone and implant intersect are captured, which allows the surgeon to fully resect all the bone required for the ukr case.It should be noted that the user had refined as well as began burring the femur and incurred a robotic drill cable disconnected error that required the user to exit and return to the case.The second time being in the case, the user was in bone refinement prior to femur bone removal.However, the model does not show the anterior edge having been refined prior to bone removal.Refer to the real intelligence cori for knee arthroplasty for refining the femur during uka.To create an accurate representation of the patient¿s operative condyle, trace the condyle surface to identify the correct amount of bone to be removed.Red bone along the medial edge of the condyle, reaching across the articular surface around the posthole was confirmed.This appears to be an entire area swept over by the bur and is possible evidence of bone tracker movement.Bone tracker movement shifts the entire virtual model from its original position defined in the checkpoint definition screen, and the bur is mislocated in relation to the new location of the virtual model.Although there are no checkpoint verification errors, if the checkpoints were taken on the bone pins/clamps/bone tracker, then tracker movement would have gone undetected because the checkpoint is still the same distance from the flat markers on the bone tracker.Refer to the real intelligence cori for knee arthroplasty for inserting checkpoint pins and defining checkpoints.Checkpoint pins are inserted in both the femur and the tibia, in positions where they will not be disturbed during bone removal.These points are collected as data points using the point probe within cori application software.Throughout the surgical procedure, the ¿checkpoints¿ are referenced to determine if either tracker has moved.This situation is captured in the risk assessment released at the time of the complaint.A review of manufacturing records indicate the software met all specifications upon release into distribution.A complaint history review found similar reports.A review of prior escalation actions was performed and found no actions that are applicable to the scope of the reported complaint.This issue will be continuously monitored through complaint investigation and post market surveillance.Based on the investigation, no containment or corrective actions are recommended at this time.Internal complaint reference number: (b)(4).Section g4 was corrected.

Manufacturer Narrative

The real intelligence cori, pn: rob10024, sn: (b)(6) used for treatment was not returned to the designated complaint unit for evaluation, therefore, visual and functional inspections could not be performed.The product was not returned however the software files were downloaded from the device and provided for investigation.The log files do not indicate any system or software issue related to the reported complaint.The screenshots were reviewed with clinical account representatives.The screenshot review confirmed the reported areas of red resection on the ¿bone removal-femur¿ screen.The red found on the anterior edge of the condyle is likely due to over-modeling of the bone.Less free collection is required when working with a smaller area of the bone, which is the case for ukr surgeries.Ukr requires point collection of the articular surface and the landmark points bounding the condyle.Therefore, with this smaller area of the bone being collected, a starter model is used to generate the bone removal model that initially appears in bone refinement, and subsequently in bone removal if the bone model had not been refined.This starter model is larger than the free collection area collected and the implant.This ensures that all areas where the bone and implant intersect are captured, which allows the surgeon to fully resect all the bone required for the ukr case.It should be noted that the user had refined as well as began burring the femur and incurred a robotic drill cable disconnected error that required the user to exit and return to the case.The second time being in the case, the user was in bone refinement prior to femur bone removal.However, the model does not show the anterior edge having been refined prior to bone removal.Refer to the real intelligence cori for knee arthroplasty for refining the femur during uka.To create an accurate representation of the patient¿s operative condyle, trace the condyle surface to identify the correct amount of bone to be removed.Red bone along the medial edge of the condyle, reaching across the articular surface around the posthole was confirmed.This appears to be an entire area swept over by the bur and is possible evidence of bone tracker movement.Bone tracker movement shifts the entire virtual model from its original position defined in the checkpoint definition screen, and the bur is mislocated in relation to the new location of the virtual model.Although there are no checkpoint verification errors, if the checkpoints were taken on the bone pins/clamps/bone tracker, then tracker movement would have gone undetected because the checkpoint is still the same distance from the flat markers on the bone tracker.Refer to the real intelligence cori for knee arthroplasty for inserting checkpoint pins and defining checkpoints.Checkpoint pins are inserted in both the femur and the tibia, in positions where they will not be disturbed during bone removal.These points are collected as data points using the point probe within cori application software.Throughout the surgical procedure, the ¿checkpoints¿ are referenced to determine if either tracker has moved.The clinical/medical evaluation concluded: ¿per complaint details, the surgeon had cut extra bone and the display showed a distorted bone model with areas of red resection following a drill disconnect error and recalibration.Reportedly, the system was showing accurate information.Although it was reported that the trackers did not move, it was communicated that the requested clinical information was not available; therefore, it is unknown if the checkpoint pin placements used for calibration were placed per user manual instructions and the clinical root cause cannot not be definitively concluded.The patient impact beyond the reported 0-30 minute surgical delay, bone overcut and subsequent slight plan adjustment to correct the resection could not be determined.No patient injury was reported, however, and the plan adjustment reportedly allowed for a correct resection with the same devices and completion of a ¿successful case¿.No further medical assessment is warranted at this time.¿ this situation is captured in the risk assessment released at the time of the complaint.A review of manufacturing records indicate the software met all specifications upon release into distribution.A complaint history review found similar reports.A review of prior escalation actions was performed and found no actions that are applicable to the scope of the reported complaint.This issue will be continuously monitored through complaint investigation and post market surveillance.Based on the investigation, no containment or corrective actions are recommended at this time.

• Brand Name: REAL INTELLIGENCE CORI
• Type of Device: ORTHOPEDIC STEREOTAXIC INSTRUMENT
• Manufacturer (Section D): BLUE BELT TECHNOLOGIES; 2905 northwest blvd ste 40; plymouth MN 55441
• Manufacturer (Section G): BLUE BELT TECHNOLOGIES; 2905 northwest blvd ste 40; plymouth MN 55441
• Manufacturer Contact: holly topping ; 7000 west william cannon drive; austin, TX 78735 ; 5123913905
• MDR Report Key: 12021769
• MDR Text Key: 257508660
• Report Number: 3010266064-2021-00453
• Device Sequence Number: 1
• Product Code: OLO
• UDI-Device Identifier: 00885556757420
• UDI-Public: 00885556757420
• Combination Product (y/n): N
• Reporter Country Code: US
• PMA/PMN Number: K201022
• Number of Events Reported: 1
• Summary Report (Y/N): N
• Report Source: Manufacturer
• Source Type: Health Professional
• Reporter Occupation: Physician
• Type of Report: Initial,Followup,Followup
• Report Date: 08/12/2022

1 Device was Involved in the Event

1 Patient was Involved in the Event

• Date FDA Received: 06/17/2021
• Is this an Adverse Event Report?: Yes
• Is this a Product Problem Report?: Yes
• Device Operator: Health Professional
• Device Model Number: ROB10024
• Device Catalogue Number: ROB10024
• Was Device Available for Evaluation?: No
• Is the Reporter a Health Professional?: Yes
• Date Manufacturer Received: 08/10/2022
• Was Device Evaluated by Manufacturer?: Device Not Returned to Manufacturer
• Is the Device Single Use?: No
• Is This a Reprocessed and Reused Single-Use Device?: No
• Type of Device Usage: Reuse
• Patient Sequence Number: 1
• Patient Outcome(s): Required Intervention