Screw placement in arthroscopically assisted osteosynthesis of radial head fractures using a reference k-wire in the radiocapitellar joint: a cadaveric study

The radial head is crucial for the stability and function of the elbow joint [1‐3]. For the treatment of radial head fractures arthroscopic-assisted techniques for fixation, in specific fracture types, are gaining popularity and present advantages compared to the classic open approach [4‐6]. Arthroscopy offers potential for better visualization of the articular surface and, with it, improved understanding of the fracture morphology. With this opportunity, a more precise anatomical reduction of the articular surface and the simultaneous treatment of concomitant injuries are potentially possible [5].

Different cadaver studies [7, 8] showed a possible exposure of the whole radial head circumference with standard arthroscopic portals for adequate screw placement. Screw placement is the crucial step in fixation of radial head fractures. As known from traumatology principles, increased primary stability of screws is achieved when they are placed perpendicular to the fracture line [9]. Most radial head fractures type I and II according to the Mason classification [10] modified by Hotchkiss [11] are partial intra-articular fractures: in these cases, axial, valgus, and external rotatory loading at the radiocapitellar joint leads to shearing of the radial head resulting in an intra-articular fracture in which in most cases the fracture line is perpendicular to the radial head articular surface, as described by Gordon et al. [12]. Hereby, the perfect screw inclination would be at an angle of 0° to the radial head articular surface.

Positioning the guide wire can be difficult during elbow arthroscopy since only a two-dimensional view of the radial head is provided and determining the radial longitudinal axis is challenging. Malpositioning of the k-wire can lead to increased surgical time, a higher amount of X-ray radiation to patient and surgeon, and at last improper screw placement. The purpose of this study was to evaluate the performance of two guiding methods and compare them to a free-hand technique of k-wire placement in a cadaveric elbow arthroscopy situation. The study hypothesis was that using a “reference k-wire” inserted in the radiocapitellar joint as a reference for “guide k-wire” placement would enable to achieve a trajectory more parallel to the radial head articular surface as compared to a free-hand k-wire placement. The possible role of an AO parallel k-wire guide in assisting k-wire placement was also evaluated.

This cadaver study was approved by the institutional review board (21-1118). For the present study, seven fresh-frozen cadaver specimens of upper extremities from human donors were used (females: 57%; right elbows: 57%; median age at death: 76 years). The specimens were evaluated for visible or radiological signs of previous trauma or deformity.

Three treatment groups were defined for subsequent comparison (techniques A, B, and C). Technique A consisted in free-hand positioning of a 2.0 mm guide k-wire in the radial head under sole visual arthroscopic control. For technique B, firstly, a “reference k-wire” was positioned percutaneously in the radiocapitellar joint and used as a reference to place the guide k-wire in the radial head. In technique C the “reference k-wire” was positioned percutaneously in the radiocapitellar joint (in the same way as in technique B) and, additionally, the AO parallel k-wire guide was used to drill the guide k-wire in the radial head.

All seven obtained specimens were allocated in a random sequence to each of the three treatment groups so that a sequence of 21 procedures was generated before the study begin. All three procedures were performed in each of the seven specimens.

Standard arthroscopy (Synergy, Arthrex, Naples, FL, USA) was performed with the elbow positioned at 90° of flexion, imitating patient positioning in lateral decubitus (Fig. 1). A high posterolateral portal was established for diagnostic arthroscopy. If necessary, an intra-articular debridement with a shaver was performed for better visualization via a standard anterolateral portal. After a single trial of positioning of the guide k-wire with the desired technique, an x-ray with a strict lateral view of the radial head and in the plane of the inserted guide k-wire was obtained with digital fluoroscopy (Fluoroscan Insight-FD Mini, Hologic Medicor GmbH, Kerpen, Germany).

After being instructed on the treatment groups and procedures, three orthopedic surgeons (RSC, DC, KW) performed each of the three techniques on the seven specimens in one sequence of 21 arthroscopies: these procedures were evaluated to assess inter-rater reliability; furthermore, to assess intra-rater reliability of the procedure, the most experienced of the three surgeons (KW) performed a second sequence of 21 arthroscopies 4 h after the first sequence. To avoid repeated perforations of the radial head at the same k-wire insertion point, the elbow was fixed in different pronation and supination positions.

The inclination angle “α” between the guide k-wire and the articular surface of the radial head was determined on digital radiographs with two decimals accuracy using dedicated software (GeoGebra 6.0.620, Hohenwarter et al., http://​www.​geogebra.​org, 2020) (Fig. 2).

Two examiners blinded to the group allocation of the radiographs (RSC, DC) analyzed all radiographs independently. Before the study, both observers were instructed about the measurement techniques. Two weeks after the initial measurement, one surgeon (RSC) re-measured all inclination angle α to evaluate the intra-observer reliability of the radiographic measurement.

Angles of 84 radiographs were obtained and analyzed. Inter- and intra-observer reliability of the measurement of the inclination angle α measured with ICC was 0.9855 (95% CI 0.9774–0.9906) and 0.9834 (95% CI 0.9673–0.9907) respectively.

With the technique A, a mean α angle of 30.1° ± SD 13 was obtained by the three surgeons; with the technique B the mean α angle dropped to 5.7° ± SD 4.5° and with the technique C to 5.4° ± 3.7 (Fig. 3). A significant difference was observed when comparing technique A to both technique B (p < 0.0001) and C (p < 0.0001). There was no difference between techniques B and C (p: n.s.). No difference was observed among the three surgeons (n.s.), neither for technique A (n.s.), B (n.s.) or C (n.s.). Inter-observer reliability measured with ICC for method A was 0.6097 (95% CI − 0.07 to 0.92), method B 0.7319 (95% CI 0.08–0.95) and method C 0.7652 (0.29–0.95). Intra-observer reliability for the second sequence of the senior author (KW) showed an ICC of 0.7781 (95% CI 0.46–0.91).

No damage to the radiocapitellar cartilage was observed arthroscopically during the study period.

The main finding of this study is that the use of an additional “reference” k-wire in the radiocapitellar joint can support to drilling the definitive guide k-wire for the cannulated screw with less variability and almost parallel to the radial head joint line with good reproducibility during arthroscopic-assisted radial head fracture fixation. A high agreement between surgeons can be achieved either by using the k-wire as a simple reference or with the additional use of the AO parallel k-wire guide.

Arthroscopy of acute elbow trauma is becoming more and more popular and has its role in the fixation of radial head fractures [13]. Arthroscopy allows direct visualization of the fracture and with this perfect reduction of this articular fracture can be verified [14]. Further advantages are reduced surgical trauma to the soft tissue compared to an open approach, the possibility to address other intra-articular soft-tissue lesions, and evacuation of loose bodies in the elbow joint. Clinical results of arthroscopic-assisted osteosynthesis are encouraging [6, 15, 16].

The techniques presented here can help surgeons place the guide wires for cannulated screw in a more parallel trajectory to the radial head articular surface (inclination angle α tend towards zero), herewith potentially enhancing the biomechanical performance of fixation and facilitating arthroscopic-assisted fracture management of radial head fractures. Surgical treatment of radial head fracture is indicated in some Mason type II, type III, and IV fractures while only Mason type II fractures so far allow an arthroscopic-assisted approach. Two-part fractures with the fracture fragment still close to the radial head articular surface are ideal candidates for arthroscopic surgery. Fragments which are dislocated outside the radiocapitellar joint usually are not amenable to being addressed by arthroscopic techniques [17].

In a cadaver study analyzing the inclination angle α of a k-wire introduced by different arthroscopic standard portals, Cucchi et al. [18] showed a mean α angle ranging from 14° to 25° obtained from the medial, 17° from the anterolateral, and − 2.6 for the more distal portal. The authors concluded a more parallel trajectory of the screw about the radial head surface is desirable to achieve superior biomechanical stability. With the here-presented additional “reference k-wire” in the radiocapitellar joint, an α angle very close to 0° (5.7° without and 5.4° with the AO parallel k-wire guide) could be achieved. Notably, the standard deviation documented in the present study using the “reference k-wire” was markedly lower (less than 5°) than the standard deviation reported both without the “reference k-wire” and that documented in the previously cited study (both 13°), suggesting a reduced variability of k-wire placement with the newly proposed technique. The application of this additional k-wire is easy, time-saving, cheap and even for inexperienced arthroscopists easily applicable. Through the soft-spot portal, or wherever the fracture line demands its placement, the k-wire can be introduced and pushed forward to the anterior elbow joint. The correct position of this wire can be assessed arthroscopically and the direction of the wire macroscopically. Care should be taken not to advance this sharp instrument too far anteriorly to avoid damage to the neurovascular structures. Moreover, the surgeon must take care, that the k-wire is not bent by the soft tissues, which would compromise the guiding effect. Further, excessive distalization of the lateral portal should be avoided and pronation of the forearm while cutting/screwing from the lateral is warranted to displace the posterior interosseous nerve medially and thus increase safety.

The high variety and the steep inclination angle obtained with the free-hand k-wire technique underline the necessity for an improved technique. Several factors add to the difficulty of free-hand k-wire placement, like soft-tissue entanglement around the k-wire, blurred vision by fracture hematoma, and resistance to placement by the annular ligament. The annular ligament reaches up to the articular surface of the radial head and is tightly fit around the radial head. To be able to place the k-wire at the correct depth to avoid intra-articular placement, the guiding k-wire must be placed at about 6–8 mm below the radial head surface. Therewith, its entry point can be blocked by the annular ligament. In a conventional technique, the k-wire needs to pull down the annular ligament to gain visibility on the entry point. In the author’s perception, this step mainly adds to misplacement and variability in the α angle. With the described techniques B and C, this step is avoided, as the “reference” k-wire dictates the direction of the guiding k-wire, which can thus be easily placed also through the annular ligament.

With the use of an additional “reference” k-wire placed in and parallel to the radiocapitellar joint, the k-wire for consecutive screw drilling can be navigated, resulting in an almost parallel position about the radial head joint line, with good reproducibility during arthroscopically assisted radial head fracture fixation.