Computer-aided planning with exact implementation into surgical technique in TKA is as accurate as intraoperative navigation

1. Introduction

As a common procedure in orthopaedic departments total knee arthroplasty (TKA) is provided for patients suffering from osteoarthritis [1, 2, 3]. To produce good clinical long-term outcomes, positioning of implants and accuracy of reconstruction of the leg axis are required [4, 5]. Several authors suggested that an alignment within 3 ∘ of varus or valgus provides the best long-term survivorship of TKA [4, 6]. Therefore, common and already successful surgical techniques (conventional) use a jig-based alignment by extramedullary (EM) or intramedullary (IM) referencing [7, 8, 9]. Of course there could be a lack of accuracy in some cases like massive deformities or compromising soft tissue [10, 11, 12, 13]. EM procedures are more liable to surgical inaccuracy and IM techniques are dependent of the femoral and tibial bone canals.

For this reason, computer-assisted navigation became more important in TKAs with its height in the last decade [14, 15, 16]. Many authors have described a significant improvement due to the mechanical axis and position of implants when a navigation system was used for alignment [17, 18, 19]. Advocates of this technique suggest that navigation in TKA could be a key technology for surgical alignment subjects [4, 20]. Nevertheless, the improvement of accuracy do not lead to superior clinical results yet [21, 22, 23].

Another tool to improve the accuracy in TKA is the preoperative planning software. Main aims of these computer programs are the correct sizing of implants and prevention of problems like femoropatellar overstuffing, a flexion-extension-mismatch with instability or postoperative pain by oversizing of the components [24]. Nevertheless, reliability and accuracy of these digital templates is sobering [25]. If a preoperative planning of the entrance location into the bone canals can improve alignment results in TKA especially for the IM techniques has not considered yet.

So the main goal of this study was to determine if the advantages of using an intraoperative navigation system during TKA can be reached by utilizing and exact implementation of a preoperative computer-aided planning software.

2. Methods

The Medical Ethics Committee of the Medical Council Westphalia-Lippe approved this study (number of ethical approval: 2016-593-f-S). Informed consent was waived due to the retrospective and observational nature of the study.

OPEN IN VIEWER

Between January 1 st 2015 and December 31 st 2016 167 patients underwent primarily TKA in a German academic teaching. During enrollment 67 patients were excluded due to the following criteria: missing pre- or postoperative leg axis measurement radiographs or missing computer added planning, malrotation of the leg axis measurement radiographs, flexion bearing of the operated knee joint during radiography, implantation of a semi-constrained or a constrained TKA. To reflect results of the power analysis a target dimension of 50 patients for each study arm was defined and data were collected continuously. Concerning this the study is based on all patients ( n = 100) hospital inclusion criteria with an equal distribution referring to the two study arms. A team of two senior surgeons, certified members of the German Knee Society ® performed all operations. Patient records in the operation room were kept electronically with a Patient Data Management System (MCC Meierhofer ® , Meierhofer AG, Munich, Germany). Data on age, sex, date, operative time, severe complications and preoperative vs. postoperative leg axis including data from the navigation system were collected retrospectively. As perioperative complications deep vein thrombosis, pulmonary embolism, cardiac infarction, deep wound infection, and death were defined.

All patients intended for TKA get a preoperative leg axis measurement by anteroposterior and lateral hip-to-ankle radiographs while standing with full weight-bearing. To transfer image data into a computer-aided planning software (MediCAD ® , HecTec GmbH, Altdorf, Germany) at a ratio of 1:1 a reference metal ball (25 mm diameter) was placed beside the knee joint center lateral or medial of the knee. During the planning process two important values were collected among others: the preoperative leg axis and the exact entrance positions for the intramedullary alignment devices in tow planes (sagittal and coronal) shown in Fig. 1. Measurements were analyzed and recorded blinded to technique allocation by a research associate who was not part of the operative team. Furthermore, two patient groups were identified: (1) conventionally operated TKA without navigation system and (2) operation with an optical navigation system (OrthoPilot ® , B.Braun Aesculap AG, Tuttlingen, Germany). The Orthopilot ® collects data via an infrared-camera-system that detects reflecting markers and uses anatomical landmarks which have to declare by the surgeon with a mobile marker. Additionally two reflecting markers were fixed at the diaphysis of the femoral and tibial bone through mini incisions. The complete navigation procedure can be assessed at the manufacture’s homepage [26].

Surgical technique was carried out through a medial parapatellar arthrotomy. All patients were treated with the cruciate ligament sacrifying nonconstrained Columbus ® knee system (B.Braun Aesculap AG, Tuttlingen, Germany). Components were fixed with bone cement (Palacos ® , Heraeus Medical, Wehrheim, Germany) after pulsed lavage and blended with a vacuum system. A straight limb axis (0 ∘ ) and a balanced extension and flexion gap was intended in both groups. To improve the alignment in the conventional treated group, bone canal entrance points from the preoperative planning (Fig. 1) were implemented with a ruler. Femoral component rotation was determined by the transepicondylar axis and 3 ∘ of external rotation relative to the posterior condyles. Tibial component was placed with 0 ∘ slope because Columbus ® knee provides 3 ∘ slope by itself using the ultra-congruent polyethylene inlay.

In summary axis values were taken from the Orthopilot ® data storage and from the planning software MediCAD ® pre- and postoperatively. Results are expressed in absolute numbers and frequencies or median and standard error of the median (SE) unless indicated otherwise. In a first step a statistical power analysis was conducted with the software package G*Power to compute the a priori required sample size. Study results are calculated with statistical software package SPSS ® Version 25 (IBM, Armonk, North Castle, New York, USA). Median differences were tested with the Mann-Whitney-U-test. For the categorical data Chi-Square test and Fisher’s exact test were used.

All patients ( n = 100) underwent primarily TKA between January 1 st 2015 and December 31 st 2016 were included. Fifty percent ( n = 50) were treated by conventional techniques and 50% ( n = 50) by navigation-assisted surgery. A balanced median age was found in the both groups with 65 (IQR: 41–80, SE = 1.49) in the conventional group and 67 (IQR: 44–82, SE = 1.25) in the navigated group ( p = 0.909). We observe 17 males and 33 females in the manual group and 10 males and 40 females in the navigated group. We compared the gender distribution of the two groups with the chi-square test and get no statistical difference between the manual and navigated group ( p = 0.176). To exclude a bias the preoperative alignment was compared between each group to ensure that both groups have the same preconditions. There is no statistically significant difference in the preoperative alignment between the groups ( p = 0.372). While there was no difference referring to the length of stay between the two groups (11d, IQR: 8–17, SE = 0.37 conventional vs. 12d, IQR: 7–27, SE = 0.57 navigated, p = 0.175), conventional TKAs could have been performed faster (80 minutes IQR: 55–154, SE = 4.14 conventional vs. 119 minutes IQR: 83–166, SE = 3.25 navigated, p = 0.000).

OPEN IN VIEWER

Main aim of this study was an analysis of the accuracy of postoperative alignment after TKA with a conventional surgical technique compared to a navigation-assisted procedure. Figure 2 shows the frequency between a resulting alignment within 3 ∘ of varus or valgus (44 in the navigated group vs. 37 in the conventional group) and out of this range (6 in the navigated group vs. 13 in the conventional group). There was no difference found between the two groups and the postoperative alignment ( p = 0.074). The distribution of the postoperative alignment of both study groups is illustrated in Fig. 3.

OPEN IN VIEWER

Furthermore, influence upon the complication rate was analyzed for both groups which is shown in Fig. 4. There was no difference ( p = 0.661).

OPEN IN VIEWER

Exact positioning of implants and accuracy of alignment are important parameters to provide an long survivorship of endoprostheses after total knee arthroplasty [4, 5]. Therefore it was suggested that a alignment within 3 ∘ of varus or valgus provides the best long-term survivorship of TKA [4, 6]. The main aim of this study was to determine, if the advantages of an intraoperative navigation system can be reached by utilizing and exact implementation of a preoperative computer-aided planning software.

Here presented findings suggest that the use of preoperative planning software and an accurate transfer into surgical procedures can provide a postoperative alignment as good as navigation-assisted surgery in TKA. It has been claimed that using a navigation system can improve accurateness of the alignment [16, 27, 28, 29, 30, 31]. In particular some authors suggest that a navigation-assisted technique could prevent higher rates of outliers [28, 29]. Nevertheless study designs (including reviews) are often heterogeneous with a lack of comparability and there is no literature that documents a clinical or functional improvement based on intraoperative navigation [27, 32]. By contrast, in the current and other studies there could not found any advantage by using a navigation system referring to an accurate alignment within 3 ∘ of varus/valgus after TKA [32, 33, 34]. Our findings suggest that a better control of the conventional alignment by preoperative computer aided planning and as exact as possible implementation of the findings into the surgical technique, in particular by find the bone canal entrance points, could be an important tool to reach results like navigated procedures.

Biasca and colleagues describe a reduction of the length of stay in hospital for navigated procedures and no difference in the operative time [33]. Both findings cannot be confirmed by data of the presented study. A longer operative time was recognized in the navigation group, while there was no difference referring to the length of stay. Other study groups confirmed our findings referring to the operative time [5, 31]. Presented results of the current study referring to severe perioperative complications do not show differences between the two study groups. These findings correlate to others in the actual literature [31, 34]. In summary, conventional and navigated procedures during TKA do not differ in their impact on the length of stay in hospital and severe complications like deep wound infection instead of a prolonged operative time for navigated techniques.

Limitations of the current study have to mentioned. Coronal plane as a parameter for alignment could not guarantee an accurate implant position over the full range of motion. Notably, like others this study cannot show differences referring to the rotational alignment [17, 26, 27]. Furthermore, a limitation of outliers could has been aroused by the fact that a skilled senior surgeon has performed all arthroplasties.

In summary, it can be stated that conventional surgical techniques in TKA are as accurate as navigated ones if an exact preoperative computer aided planning is implemented during surgery. Probably, an important tool to reach this accurateness is the femoral and tibial intramedullary alignment, which is able to reproduce planning parameters easier than during extramedullary procedures. Advantages of conventional surgery like faster surgery duration, no investment and maintenance costs for the navigation system, and no additional incisions or drill holes should be mentioned, too.