A quantitative analysis of two cementing techniques in total elbow replacement: A cadaveric study

Abstract

One of the reasons for failure of total elbow replacement is loosening of the ulnar component. Cementing techniques are often outdated. A special small nozzle is needed for the ulnar component, when a cement gun is used. This may not always be available, or surgeons may prefer to use a syringe. We postulated that the use of a cement gun and smaller nozzle would result in improved filling of the ulnar canal. A cadaveric study was performed in which the ulnas of paired specimens were cemented with a cement gun or with a syringe. A 3D printed ulnar component was inserted and computed tomography scanning was performed on all specimens. Filling of the ulnar intramedullary canal was analysed using 3D reconstructions of the specimens. A greater degree of filling was seen by the use of the cement gun in 85.7%. Filling was 52.7% in the syringe group (25.1–78.7%), compared to 63.3% for the cement gun group (p < 0.05). The use of a small nozzle cement gun provided a significantly higher filling degree of the ulnar canal. We recommend to always use a cement gun with a specific small nozzle to cement the ulnar component in total elbow arthroplasty.

Introduction

The indications for total elbow arthroplasty (TEA) have changed from predominantly rheumatoid arthritis to acute trauma and post-traumatic indications.^1–4

As these patients are often younger and more demanding inevitably survivorship of implants will be challenged more than ever.^4,5

Despite the advances in the design of TEA, the survival rate of TEA remains significantly lower compared to total hip and knee arthroplasties,^6–8 and the overall complication and revision rate have been reported to be as high as 34%. ⁵ Besides deep infection, bushing wear and component fracture, aseptic loosening is one of the main causes of failure in TEA. ⁵ Radiolucent lines are commonly found on follow-up and the incidence of loosening has been reported to be up to 15%.^4,9

The aetiology of aseptic loosening is still not completely understood. Prosthetic design and material properties play a role, as well as procedures used during the insertion of the components and the lack of connectivity at the interface between the implant and surrounding cement or between the cement and bone. ¹⁰ In TEA, aseptic loosening has been shown to result from primary failure of the bone–cement interface or polyethylene wear which may induce secondary failure of the bone–cement interface.^4,9 Aseptic loosening in TEA is more likely to involve the ulnar component.^11–14 Anterior impingement causing traction on the components is another potential mechanism for loosening. ¹¹ The use of porous-coated ulnar components with high resistance against pull-out forces, in combination with the correct technique of implantation, could decrease this mechanism of failure. ¹¹

Inadequate cementing technique will contribute to loosening^6,15 and is a potentially preventable cause of implant failure. ⁹ Analysis of the fibrous tissue, bone–cement interface, membrane in failed cemented total hip prosthesis showed polymethylmethacrylate debris, due to motion of the implant. This can induce an inflammatory response with macrophage, giant cell reaction and foreign body granulomas, leading to osteolysis and prosthetic loosening.^10,14 High shear loading on a thin edge of the polyethylene bushing in linked semiconstrained TEA most likely increases wear debris and might explain the slightly higher rate of combined clinical and radiographic aseptic loosening in comparison with the unlinked designs. ⁹ It has been shown that decreased thickness of the cement mantle will cause micromotion. ¹⁶ Shear strength at the bone–cement interface has been found to be linearly dependent on the depth of penetration of the cement into the bone. ¹⁷ Penetration of cement into the bone can be improved by the use of distal bone plugging, pressure lavage and pressurized retrograde insertion of low-viscosity cement.^17,18

Cementing technique of both the humeral and the ulnar components of TEA have not been well developed. While the use of a cement gun is well established in hip or knee replacement surgery, there are few reports on cementing techniques in TEA. A special narrow nozzle is needed for the ulnar component when a cement gun is used as a standard nozzle does not fit in the ulnar canal. This may not always be available, or surgeons may decide it is not necessary and prefer to use a syringe. The current study compares the degree of cement filling when cementing is performed either with a syringe or a cement gun.

The hypothesis of this study is that the use of a cement gun and smaller nozzle will result in improved filling of the ulnar canal, when compared to using a syringe.

Materials and methods

Fourteen fresh-frozen cadaveric ulnas, bilaterally obtained from four male and three female subjects, were tested in a paired study design to evaluate the cement filling degree of two cementing techniques. Specimens were thawed at room temperature and soft tissue was removed. The ulnar bone was prepared according to the surgical guide of the Conrad-Morrey prosthesis (Zimmer Biomet, Warsaw, IN, USA).

The ulnar components were three-dimensionally (3D) printed in polyamide (Figure 1). For practical reasons, the ulnar implants were printed in one size only (small) to fit all specimens.

Figure 1.

The ulnar components, three-dimensionally (3D) printed in polyamide. (Courtesy of MoRe Foundation.).

The intramedullary canal was occluded with a bone plug used as a cement restrictor, irrigated and the bony surface was dried to remove redundant fat and blood. Simplex P cement (Stryker, Kalamazoo, MI, USA) was mixed at room temperature (21℃) and atmospheric pressure using an open bowl with a spatula. The cement was delivered in a low-viscosity state, 2 min after the start of the mixture, using either a 60 mL syringe or a specific cement gun and nozzle (Zimmer Biomet) until the canal was considered to be filled completely. The canal was not vented and, other than the bone block, no pressurisation techniques were used. The 3D printed models were implanted at the required depth.

Computerized tomography (CT) images were obtained (0.625 mm slice thickness, 0.312 mm slice increments, 49 mA, 100 kV) and analysed using 3D modelling. Segmentations were performed by one investigator and consequently reviewed by two blinded evaluators. Pilot studies investigating interobserver variability revealed no significant difference between three investigators.

Accurate 3D models of the intramedullary space and cemented volume around the stem of the ulna prosthesis were generated (Mimics Medical or Mimics InPrint v20.0, Materialise NV, Leuven, Belgium).

The filling degree was calculated as the ratio of the cemented volume to the volume of the intramedullary canal. As limits for these computed volumes, a plane was drawn at the tip of the ulna prosthesis perpendicular to the anatomical axis of the ulna and a second plane, parallel to the first plane, through the coronoid process (Figure 2).

Figure 2.

Intramedullary space (green), cement (yellow) and ulnar component (red) were marked on 2D CT slices and converted into 3D models of the ulna (blue). The filling degree is calculated by the subtraction of the cemented volume from the volume of the intramedullary space. (Courtesy of MoRe Foundation.)

A comparison in filling degree was made between the two cementing techniques (Figure 3).

Figure 3.

(a) Comparison of intramedullary filling was done on paired specimens. This specimen showed an 84% filling when using a cement gun. (Courtesy of MoRe Foundation.) (b) Filling of 78.7% was found in the paired specimen when using a syringe respectively. (Courtesy of MoRe Foundation.)

Results

A greater degree of filling was seen by the use of the cement gun in 85.7% of the cadaveric subjects (Figure 4).

Figure 4.

Paired differences in cement filling degree (%) between the cement gun and syringe. The line represents equal filling degree for both techniques. Dots above the line show greater filling degree with the cement gun technique and dots below the line show better filling degree with the syringe technique. The cement gun technique was superior in all but one pair. (Courtesy of MoRe Foundation.)

The average filling degree for the syringe group was 52.7% (25.1–78.7%), compared to 63.3% (40.1–84%) for the cement gun group (Figure 5). This difference was statistically significant (p < 0.028, related-samples Wilcoxon Signed Rank test).

Figure 5.

Filling degree (%) in the syringe group and the cement gun group. (Courtesy of MoRe Foundation.)

The average filling degree for men and women was respectively 43.7% and 64.6% for the syringe group, and 53.3% and 76.5% for the cement gun group. A significant difference was noticed in the overall degree of filling, with an average of 45.7% for men and 70.8% for women (p = 0.043, independent-samples Mann Whitney U test).

Furthermore, a correlation was found between the proportion of the intramedullary canal and the filling degree (Pearson correlation test). Less cement coverage was found in specimens with a larger intramedullary canal.

Discussion

We hypothesized that implementation of cementing techniques, similar to those used in total hip and knee arthroplasties, would lead to an improvement in cement coverage of the ulnar component, which may contribute to a stronger initial and long-term fixation and may prevent wear debris to access the cement–bone interface, potentially decreasing the incidence of aseptic loosening of the ulnar component.

The advanced cementing technique in the current study included: canal plugging, irrigation and drying to remove blood and fat and low-viscosity cement. Deliverance of cement was performed with a syringe or with a cement gun and special nozzle.

The use of a cement gun produced a significantly greater cement filling. This result was comparable with the results of Faber et al., who reported, in addition to a better filling degree, a significant difference in the load to failure and the failure stress in the advanced group, comparing two cementing techniques in the distal humerus. ⁶

MacDonald et al. reported an increase of 52% in interfacial shear strengths when distal bone plugging, lavage and cement pressurisation were used in a femoral canine model. ¹⁷

A weakness of our study was the use of a bone block as a cement restrictor, as the block may have shifted during cement insertion. We therefore chose to measure filling from the tip of the prosthesis instead of to the most distal tip of the cement. We did not study the effect of type of cement used, cement preparation and mixing technique and of the use of antibiotic mixture on coverage of the ulnar component. Although these may be important parameters, we focussed on the differences in mechanical delivery of the cement with a cement gun or syringe. All other parameters were identical between the groups tested.

Another weakness of our study was the use of one single size of ulnar component, irrespective of the size of the ulna but given the paired study design the potential effect on the results will probably be minimal. We found a correlation between the proportion of the intramedullary canal and the filling degree of cement. Less cement coverage was found in specimens with a larger intramedullary canal. This finding suggests that, besides advanced cementing techniques, using the correct size of implant positively influences the filling degree of cement in TEA.

Conclusion

Aseptic loosening of the ulnar component is a potentially preventable reason for failure of the implant. This study shows that the use of a cement gun and appropriately sized nozzle improves the cement mantle and filling of the intramedullary canal.

ORCID iDs

Matthias Vanhees https://orcid.org/0000-0001-6905-9092

Roger van Riet https://orcid.org/0000-0002-4866-5867