Application of Digital Surgery in Orthopedics: THA and TKP

Introduction

The application of robotic surgery is widespread in orthopedics surgery, especially in “total hip 1. In digital surgeries of THA and TKP, computer navigation systems are used for pre-operative planning. The data obtained is then integrated into a surgical system where robots assist the surgeon during the operation.

Robotic surgery is widely accepted by the orthopedics community for two reasons. First, the technology is well-suited to operations on bones. Since bones, unlike soft-tissues, are less-prone to deformation when being pressed on, computer navigation and mapping of bones, which use software based on the rigid-body assumption, are accurate and efficient. Furthermore, robotic surgery increases the accuracy of implantation and component placement in THA and TKP. Current surgical techniques often result in inaccurate placing and balancing of hip replacements, knee components, or soft-tissues. With computer navigation system and robotics assisted surgery, precision and accuracy is greatly improved; this in turn leads to better long-term outcomes², ³.

http://www.orthopilot.com/

In this section, we’ll present the science and technology behind the computer navigation and robotics system as well as their implementation in the clinic.

Computer Aided Orthopedic Surgery

A. Surgical Navigation Systems

Surgical navigation systems allow the surgeon to perform surgical actions in real time using information conveyed through a virtual world, which consist of computer-generated models of surgical instruments and the virtual representations of the anatomy being operated on. Virtual representations can be generated from data obtained through Computer Tomography scans, Fluoroscopy, or Image-free navigation. The concept of surgical navigation system is similar to a real time GPS vehicle tracking system in which the location and movement of the vehicle is constantly displayed on a map⁴.

The navigation systems currently in use can be characterized by three major components: the surgical object, the virtual object, and the navigator. The surgical object is the bones and accompanying tissues in the surgical field. The virtual object is the virtual representation of the surgical object. Finally, the navigator establishes a coordinate system in which the location and orientation of the target as well as “end-effectors” are expressed. The “end-effectors” can be surgical instruments or active devices⁴.

^{4, S-A7}

Three major procedural requirements are essential to successful navigation. First, end-effectors must be calibrated for correct representation of their shapes and geometry in the coordinate system established by the navigator. Second, “registration” establishes correspondence between the surgical and the virtual object, which is essential to the display of the end-effectors’ locations in the virtual representation. Finally, “dynamic referencing” using dynamic reference bases establishes a local coordinate system that compensates for possible motion of the navigator or the surgical object during surgical action. Examples of dynamic reference bases are optical markers⁴.

Components	Definition	Procedures	Definition
Surgical Object	The anatomy to be operated on.	End-Effectors Calibration	To establish correct representation of the shapes and geometry of surgical instruments/devices in the coordinate system
Virtual Object	The virtual representation of the surgical object, usually generated from data obtained by CT-scans, fluoroscopy, or image-free tracking system	Registration	Established correspondence between the surgical and the virtual object through identification of structures on the bony surface and corresponding features in the image data⁴.
Navigator	Established a coordinate system for the accurate correspondence between the surgical and virtual object	Dynamic Referencing	To establish local coordination system to compensate for possible motion of the navigator and/or surgical object during the operation

Table 1. Major components and procedures of Computer Navigation Systems

Current navigation systems can be sorted into CT-based, fluoroscopy-based, and image-free based. The navigation systems differ in the way information on the surgical object is acquired.

i) CT-based surgical navigation

CT-scan is utilized to acquire data on the surgical object pre-operatively. The data is then loaded into a navigation system. During the operation, the CT-images must be correlated with the patient’s anatomy through registration. CT-based navigation systems have the advantage of providing detailed 3D images. However, CT-scan exposes the patient to a considerable amount of radiation, and errors in registration pose a possible pitfall for the operation ⁴, ⁵.

ii) Fluoroscopy-based surgical navigation

Mobile fluoroscopic devices, also called C-arms, provides real-time feedback on the surgical objects and the end-effectors. Patients are placed in the “C” of the C-arms during the surgery (see picture below). Since 1999, 3D fluoroscopy-based surgical navigation is made available. The 3-D image is reconstructed from a series of 2-D images taken in different orientations. Fluoroscopy has the advantage of providing real-time movements and changes of the surgical objects as well as the end-effectors. Changes in the anatomy of the patient due to surgical actions can be visualized. However, the C-arm must be carefully calibrated to ensure the accuracy of the acquired images. In 3-D fluoroscopy, any patient movements, including respiration, can corrupt the geometrical correspondence of the dataset with the actual anatomy. These potential pitfalls must be carefully avoided for fluoroscopy-based surgical system ⁴, ⁵.

Basic Principles of CAOS, S-A11, Nolte and Beutler

iii) Image-free surgical navigation

The basic concept of image-free surgical navigation is to create virtual representation of the surgical object using a tracking system defining various anatomical structures⁴. No pre- or intra-operative radiological images are used, instead, the virtual representation is “surgeon-defined”⁴. There are various image-free navigation systems, with the most-advanced forms utilizing “bone-morphing technologies” ⁴, ⁶. While image-free navigation renders errors present in CT-based and fluoroscopy-based navigation void, the surgeon-defined virtual representation is subject to other pitfalls. The accuracy of the virtual representation cannot be verified with recorded information unless it is grossly inconsistent⁵. Moreover, atypical anatomy may not be accurately represented by a generated representation based on sparse recording data on the bone surface ⁵.

Types of Navigation	CT-based	Fluoroscopy-based	Image-free based
Devices used	CT scanner	Mobile Fluoroscopic devices (C-arm)	Varies, including anatomy landmarks and statistical models.
Advantages	Allow pre-operative planning on detailed 3D images	Real-time images of the surgical field.	No errors similar to those of CT-based and Fluoroscopy -based systems
Disadvantages	High does of radiation on patients, errors in registration can lead to disastrous result	C-arm calibration errors and minute patient movements during scanning can corrupt images	Hard to verify the accuracy of surgeon-defined representation, atypical anatomy poses a problem

Table II. Comparing types of Navigation systems

B. Robotic Systems

Accompanying the computer navigation systems are the corresponding robotics that actively or semi-actively participate in surgery. Robots are an integral component of robotic surgery. The pre-operative and intra-operative data obtained through computer navigation is essential for the performance of the robot. This section contains information about the interaction modes between the surgeon and the robot as well as examples of current robotics systems.

i) Interaction Modes:

A robot can operate with complete autonomy. In this case, the robot carries out the pre-operative plans without immediate human intervention. One such example is the use of robots in hip joint replacement in which the robot carries out the complex bone-milling process with a precision unattainable by humans¹.
A robot can assist the surgeon. In this case, the surgeon and the robot interact with each other during the operation. One such example is a robot that provides “active constraint”¹.

The concept of “active constraint” imposes four regions on the surgical object⁷.

Safe region- a region away from soft tissues and free motion is permitted
Close region- the surgeon is near to soft tissue and motion is partially constrained by the robot.
Boundary region- the surgeon is about to touch soft tissue and motion is severely constrained.
Forbidden region- the surgeon is cutting into the soft tissue and motion is not allowed.

http://babbage.me.ic.ac.uk/case/mim//

For example, a surgeon maneuvers the cutting tool of at the end of a robot manipulator while the robot monitors the surgeon’s motion and permits or prevents free motion depending on the location of the surgeon’s action. The robot is equipped with sensors and possesses registered pre-operative information to determine whether the surgeon is operating at the right spot.

C) The third interactive mode falls under the category of teleoperation.

In this case, the surgeon controls every motion of the robot through the control console. Information from the surgical field is transmitted to the surgeon for decision-making. This type of interactive mode is common in minimally invasive surgery, and it renders long-distance operation possible.

“Robotics For Surgery” Annu. Rev. Biomed. Eng. 1999.01 211-240, Robert D. Howe and Yoky Matsuoka

Commercial Computer Aided Orthopedic Surgery System

A) RoboDoc surgical system - The Robodoc surgical system consists of the Orthodoc, a pre-operative planner, and the RoboDoc, a surgical tool. The RoboDoc was developed by Integrated Surgical Systems in the 1980s and is currently used in hospitals in Europe. It is first used in total hip replacements and its use is now extended to revision hip replacement and total knee replacement¹.

The RoboDoc surgical system uses Computer Tomography to obtain structural information of the surgical object pre-operatively. The OrthoDoc allows the surgeon to construct a pre-operative bone-milling procedure. Before the bone-milling procedure, registration is conducted to match the pre-operative data with the physical location of the surgical object. In total hip replacement, RoboDoc is used to cut the femoral cavity precisely.

RoboDoc (http://www.robot.md/publications/robodoc.pdf)

During the surgery, the surgeon controls the robot via a hand-held terminal. The robot is manually guided to the desired location by the surgeon. Once the end-effector is brought to the designated location, the RoboDoc cuts the bone automatically. The surgeon can see inside the bone and monitor the robotic cutter’s position within the bone through a “real time monitor”. Once the cutting is completed, the surgeon resumes the normal manual procedure³.

The RoboDoc has two main advantages over manual procedure. First, the femoral pocket is more accurately formed. Second, the pre-operative planning with CT-scan data allows the surgeon to optimize the implant size and placement for each patient¹.

Nevertheless, there are also two disadvantages inherent in the RoboDoc system. First, the registration process requires the use of registration pins on the bone surface. A traumatic pin-placing procedure and a slow pin-finding registration process are required. Second, a complex and time-consuming method is used to fix the femur to the base of the robot¹.

While the RoboDoc system is still awaiting FDA approval in the US, the system is used in more than 35 hospitals in Europe. It takes three weeks to three months for doctors to feel comfortable with the system³, and patients are positive to the use of the robot. The cases involving RoboDoc result in one-third less hospital staying time³.

One study estimates that the complete RoboDoc system, including a computer-based planning station, costs USD 600,000. Another study suggests that a RoboDoc surgery costs USD 700 more than conventional surgery due to additional operation room time (9).

RoboDoc (http://www.robot.md/publications/robodoc.pdf)

B) Acrobot- Developed by Professor Brian Davies and colleagues, Imperial College London, the Acrobot is designed for Total Knee Replacement. The robotics system uses CT-scans to obtain pre-operative data for pre-operative planning. The Acrobot system employs the concept of “Active Constraint” (see above). During the operation, the surgeon stays in control, but the robot applies resistive force to the surgeon’s hand based on the defined surgical region.

This system has several advantages over manual procedures. First, the use of “active constraints” confines the surgeon’s actions to the safe region. Second, the Acrobot system assists the surgeon to achieve accurate cuts and trajectories while ensuring the pre-operative plan is achieved⁸.

As of April, 2005, the Acrobot surgical system is not yet approved by the USFDA.

No comprehensive economic analysis is available for the Acrobot system.

http://www.lirmm.fr/manifs/UEE/docs/slides/Davies.pdf

C) CASPAR- Computer Assisted Surgical Planning and Robotics is an integrated system developed by the German company OrtoMaquet¹⁰. The system is mainly used for knee surgeries.

http://www.lirmm.fr/manifs/UEE/docs/slides/Perrier1.pdf

A CASPAR procedure consists of four distinct phases. First, a surgery is performed to insert rigid bodies, commonly pins, at different locations around the knee. The procedure requires local or total anesthesia, and a tibial pin is placed 8 to 10 cm above and below the knee joint. This procedure is required for registration and referencing. Second, CT-scan is performed to retrieve pre-operative data and images of the knee. The surgeon can then perform pre-operative planning with the data. It generally involves measuring the femur and the tibia and planning the femur and tibia tunnels in a 3D environment. Finally, the robotics can do the tunnel drillings automatically under surgeon supervision¹¹.

In one study, the average surgery time increases by thirty minutes. The average registration errors are 0.3mm for tibia and 0.6mm for femur. The degree of precision is found to be 0.5mm¹¹.

The CASPAR system offers precise tunnel drillings that cannot be done as accurately with human hands. Also, the CT-scan allows pre-operative planning and gives surgeons the chance to optimize their operation. Nevertheless, the system suffers the disadvantages of increased cost, increased surgery time, and an additional surgery for pins implementation¹¹.

D) CRIGOS - Compact Robot system for Image-Guided Orthopedic Surgery

CRIGOS is an ongoing project by the biomedical community to build small and cost-effective robots designed specifically for surgery. Current systems such as RoboDoc and CASPAR use robots based on industrial manufacturing robots. Scientists and engineers involved in the CRIGOS project aim to design specific small, cheap and sterile robots as well as the accompanying software and pre-operative navigation systems for specific surgeries. Certain prototypes have been developed, but the project is still under development¹³,¹⁴.

http://www.ame.hia.rwth-aachen.de/research/cht/Crigos1.html

The advantages and disadvantages of robotic systems are listed in the table below:

“Robotics For Surgery” Howe and Matsuoka

Computer Aided Orthopedic Surgery is advancing quickly and becoming more widely accepted. With the improvement of the digital surgical systems, CAOS has a bright future in the medical field.