How Robotic Spine Surgery Works: Planning, Accuracy, Safety

Robotic spine surgery works in three stages: a 3D scan of the spine is taken in theatre, every screw is planned on screen for entry point, trajectory, diameter and length, and a rigid robotic arm then guides the drill along that plan. The surgeon still performs the decompression and correction; the robot standardises where the metal goes.

Dr. Puneet Girdhar, head of spine surgery at BLK-Max Super Speciality Hospital in New Delhi, defines accuracy in spinal instrumentation with a dartboard. Avoiding damage with a screw is not the standard he means. The standard is the bullseye: every screw placed exactly on the planned trajectory, every time. Across his last 350 cases with navigation and robotics, he reports, that is what his team has achieved, with 100 per cent accuracy. For three years he has used robotics on an everyday basis for the hardware stage of his operations, the placement of screws and cages, and he credits it with saving time and making his instrumented surgery more minimally invasive and more precise.

This article sets out how a robotic spinal operation is planned and executed in his theatre, why the technology changes the safety equation for revision and upper cervical surgery, and where the surgeon's own skill remains indispensable. The wider case for technology-supported surgery is made in the complete guide to advances in spinal surgery.

How robotic spine surgery works, step by step

A robotic spinal operation follows the same sequence every time: fix a reference frame, acquire a 3D image, plan the hardware virtually, then execute the plan with the robotic arm under real-time imaging.

Setting up: the reference frame and the 3D spin

Every robotic case in Dr. Girdhar's practice begins the same way. A reference frame is fixed to the patient's lower back, anchored in the iliac crest. The frame gives the system a fixed point against which the position of the spine and of every instrument can be calculated.

A 3D C-arm, a mobile X-ray unit that rotates around the operating table, then spins around the patient and acquires a three-dimensional image of the spine. The robotic software displays that image in four panels:

• Sagittal, the side view.
• Coronal, the front view.
• Axial, the cross-section through each vertebra.
• A 3D reconstruction, which can be rotated on screen.

The 3D model is the planning surface. The surgeon can virtually pick up the spine, turn it, and inspect it from any angle before a single instrument enters the body.

Planning before cutting

The planning stage is where robotics departs most clearly from freehand and conventional fluoroscopic technique. Using the software, the surgeon drags and drops the hardware onto the spine model: screws, cages and rods, arranged the way the final construct should look. For each screw the plan sets:

• The entry point.
• The trajectory.
• The diameter.
• The length.

The whole construct is designed on screen and only then executed. Dr. Girdhar regards this as the step beyond conventional navigation. Navigation shows the surgeon where an instrument is as the operation proceeds. A robotic system with planning software lets the surgeon design everything in advance, so the operation becomes the execution of a plan rather than a sequence of decisions made on the go.

Execution: the arm and the real-time overlay

With the plan complete, the robotic arm is brought into action. The surgeon gives a command, for example an L4 screw on the right, and the arm aligns itself along the planned trajectory.

As drilling begins, the screen shows the instrument's position in real time. In a minimally invasive operation this matters, because the surgeon cannot see inside the body through a small incision. Everything the surgeon sees of the screw path is on the screen. In Dr. Girdhar's display the planned screw appears in blue and the screw being executed in yellow. When the two overlap in every view, the screw is going exactly where it was designed to go.

The software reports further detail in real time: the length and diameter of the screw, the type of screw chosen, any deflection from the planned path, and the trajectory angle itself. In one case he showed, the display read a trajectory of 27.7 degrees. The same workflow applies to multi-level constructs.

Mirror-image screws

The post-operative images carry a signature. Screws placed on either side of a vertebra appear as mirror images of one another, with identical trajectories. Dr. Girdhar's observation is that this symmetry occurs only with robotics. It is a visible measure of the consistency the technology brings.

Radiation exposure and operating time

Robotic spine surgery reduces radiation to a single 3D scan and shortens the hardware stage of the operation. Conventional fluoroscopic screw placement depends on repeated X-ray shots or repeated C-arm spins to check position. With robotics, the only radiation exposure is the single 3D spin at the start, taken while no one but the patient is in the theatre. For the patient it is a one-time dose. For the surgeons and theatre staff the procedure is effectively radiation-free, a benefit that compounds for a team operating every day.

The same feature saves time. Without the need for repeated shots and spins, and with the plan already made, the hardware stage moves quickly. Dr. Girdhar sees the time saved as one of the main gains robotics has given him.

Expandable cages and restoring the lumbar curve

An expandable cage restores disc height and lumbar curvature in a collapsed segment. In a fusion where the disc space has collapsed, he now often uses expandable cages. The cage is inserted in its compact form and then expanded within the disc space. This restores disc height and, with it, the natural curvature of the lumbar spine, returning the segment to the alignment the patient had before the pathology developed. Expandable cages also help reduce slips: in patients with a grade two listhesis, Dr. Girdhar achieves an easy reduction with good disc height and good segmental as well as global sagittal balance. The minimally invasive fusion this hardware belongs to is described in the minimally invasive surgery article.

Robotic revision spine surgery: planning from safe zones

Robotics offers its clearest advantage in a patient who has been operated on before. In revision surgery the normal bony landmarks may be missing, and scar tissue covers the field. Working freehand, the surgeon must find the landmarks while avoiding scarred areas, and there is a risk of entering the dura.

With robotics, the surgeon does not need to find the landmarks in the wound. The plan is made on the 3D image, from safe zones, before the operation starts. Dr. Girdhar illustrated this with a patient who had previously undergone a laminectomy. Only small areas of bone remained available for safe screw placement, and the plan was built to use those areas without straying into the previously operated zone. The result was a transforaminal lumbar interbody fusion (TLIF) performed in a revision setting through two small incisions, with mirror-image screws and a cage.

Robotic C1-C2 transarticular screw fixation

Robotics turns a C1-C2 transarticular screw, placed in one of the most demanding regions of the spine, into a fully planned procedure, which in one case took half an hour. The atlantoaxial joint, between the first and second cervical vertebrae, is among the most demanding regions for screw placement. The bony elements are small and their surfaces slippery, so a drill can skid or skive across the bone. Placed conventionally under C-arm imaging, a C1-C2 transarticular screw was a difficult operation.

Dr. Girdhar's team performed a robotic C1-C2 transarticular screw fixation in a patient with an upper cervical fracture. The entire planning was done on the robotic software, and the screw was driven with a power drill along the planned trajectory through a rigid robotic arm. The software shows the landmarks, so the surgeon does not have to find them. The procedure took half an hour. In his estimate, an open technique, with the time spent exposing and identifying landmarks, would have taken two to three hours. The post-operative images showed the screw passing through the centre of the C1-C2 joint, as planned.

Limits of robotic spine surgery: what the robot does not do

Dr. Girdhar draws a clear line around the technology. Robotics standardises implant placement, and it allows a team to place hardware with the same precision regardless of each surgeon's years of experience. It does not decompress a nerve, reduce a slip, perform a deformity correction or achieve sagittal balance. Those remain matters of the surgeon's skill and judgement. His argument is for handing over to the technology the work that can be handed over, so that the surgeon's attention is concentrated on the parts of the operation that only a surgeon can do.

The same thinking runs through the rest of his operating environment. Intraoperative 3D imaging, multimodal neuromonitoring and ultrasonic bone cutters each take a source of variability or risk out of the surgeon's hands. In the most complex deformity corrections, where an entire vertebra is removed around the spinal cord, these tools work together, as the scoliosis and deformity article shows.

Why robotic spine surgery matters to the referring doctor

For the doctor deciding whether to refer, robotics changes two things. First, it lowers the risk in placing metal close to the spinal cord and nerves. Second, it makes minimally invasive instrumentation feasible in situations that once demanded an open operation: revision surgery after a previous laminectomy, higher-grade slips, and the small, difficult anatomy of the upper cervical spine. A patient who has already had one open operation may now be a candidate for a small-incision revision planned entirely in advance.