Surgical treatment of Neer type II and type V lateral clavicular fractures: comparison of hook plate versus superior plate with lateral extension: a retrospective cohort study

A retrospective cohort study was performed using data from two level II trauma centers. All patients with an unstable LCF who were treated operatively between January 2011 and June 2016 were eligible for inclusion. Inclusion criteria were: (1) acute LCF, (2) age 18 years or older, (3) Neer type II or type V fracture, (4) fixation with CHP or SCPLE, (5) fixation within 2 weeks of injury, and (6) minimum of one-year follow-up. Exclusion criteria were: (1) history of prior shoulder injuries or (2) neurovascular disorders of the affected shoulder. Data collection was performed by reviewing electronic medical records, operative reports, radiology reports, and telephone interviews by an independent research fellow. Electronic medical records were reviewed to collect baseline characteristics regarding affected shoulder, age, gender, trauma date, trauma mechanism, time from injury to surgery, fixation method, previous shoulder injuries, and lateral fragment size. Lateral fragment size was measured in millimeters (mm) on the anterior–posterior view radiograph. Overall lateral fragment size was defined as the total length of the largest lateral fragment. The largest intact bicortical fragment, which would allow for adequate screw fixation, was considered as the bicortical lateral fragment length. Informed consent was obtained from all subjects, and approval was granted by the institutional review board.

Patients were treated by means of open reduction and internal fixation (ORIF) using a CHP (3.5 mm LCP; Depuy Synthes GmbH, Oberdorf, Switzerland) or SCPLE (3.5/2.7 mm LCP; Depuy Synthes GmbH, Oberdorf, Switzerland). Implant selection was based on the surgeon’s preference. CHP and SCPLE fixation were performed by several surgeons in both trauma centers. Operations were performed under general anesthesia with the patient placed in a beach chair position. An incision was made using a standard superior approach. The fracture site was exposed preserving as much periosteum as possible. Reduction was performed under direct visualization, and fragments were temporarily fixated using K-wires or reduction forceps. Fracture reduction, implant position, and screw placement were checked under fluoroscopic guidance. Coracoclavicular ligament repair was not routinely performed. Finally, the fascia and skin were closed in layers.

In cases of CHP fixation, a small incision was made in the posterior capsule of the acromioclavicular joint to allow sub-acromial hook placement. Trial plates were used to determine correct length and depth. Definitive CHP fixation was completed with the insertion of 3.5 mm angular stable or conventional screws (Fig. 1).

In cases of SCPLE fixation, there was no involvement of the acromioclavicular joint. A plate with an appropriate length was chosen to allow adequate fixation with 3.5 mm conventional or angular stable screws in the medial fragment and smaller 2.7 mm angular stable screws in the lateral end (Fig. 2).

Both groups received the same postoperative management. Radiographs were taken 1 day postoperatively. Patients were temporarily immobilized in a sling until the pain subsided; early mobilization and active range of motion exercises were allowed when tolerated. Weight-bearing activities and resisted exercises were not permitted until approval from the treating surgeon. Follow-up visits were scheduled at 2, 4, and 12 weeks postoperatively. Additional outpatient visits were scheduled depending on fracture consolidation. Removal of the SCPLE was not routinely performed, as opposed to the CHP where removal was recommended to all patients.

Functional outcome was assessed at least 12 months following ORIF, using the Dutch language version of the QuickDASH score. The QuickDASH is a validated and shortened version of the Disabilities of the Arm, Shoulder and Hand questionnaire (DASH). The QuickDASH is a patient-reported outcome instrument developed to measure upper extremity disability and symptoms, resulting in a score ranging from 0 (no disability) to 100 (most severe disability) [18, 19].

Secondary outcomes were the numerical rating scale (NRS) pain score at rest and during activity, complications, revision surgery and implant removal. The NRS is a reliable and commonly used 11-point scale to measure pain intensity, ranging from 0 (no pain) to 10 (worst imaginable pain) [20]. Complications included infection, non-union, mal-union, implant failure, and implant removal-related complications. Infections were subdivided in superficial-skin or deep-wound infection. Superficial infection was defined as redness, swelling, or purulent discharge from the wound that was treated with antibiotics alone. If surgical irrigation and debridement was required, it was considered a deep infection. Non-union was defined as the absence of fracture consolidation 6 months after surgery. Mal-union was defined as a symptomatic deformity of the clavicle. Implant failure was defined as implant displacement, implant breakage, or breakage of screws. Revision surgery was defined as the need for subsequent surgery other than implant removal. Infection and re-fracture following implant removal were considered implant removal-related complications. Implant-related irritation and indication for implant removal were analyzed using a series of questions developed by Hulsmans et al. [21]. Responses to these questions allowed categorization of implant removal into (1) routinely or on patient’s request without irritation or (2) patient’s request due to irritation. Patients with the implant still in situ received a different series of questions, leading to categorization of why implant was not removed; (1) not experiencing irritation, (2) experiencing irritation but removal not necessary, (3) experiencing irritation but no request for removal due to fear of re-operation, or (4) experiencing irritation, considering removal.

Descriptive results are presented as mean values with standard deviations and range (± SD, range) and median values with interquartile range (IQR) or absolute numbers and percentages (%). Continuous variables were evaluated using an independent sample t test or Mann–Whitney U test. Categorical variables were compared using the Pearson’s Chi-squared test. The Fisher’s exact test was used in case of small count sizes. Mean differences and relative risks (RR) were calculated with 95% confidence intervals (CIs). The significance level was defined as a p value < 0.05. All statistical analyses were performed using IBM SPSS Statistics version 24 for Windows (IBM Corp, Armonk, NY).

A flowchart of the patient cohort is shown in Fig. 3. In total, 76 patients met the inclusion criteria. However, eight patients could not be contacted, and one patient refused participation. This resulted in the inclusion of 67 patients (88%) for analysis. The baseline characteristics are shown in Table 1. The CHP group included 19 patients (28%) compared to 48 patients (72%) in the SCPLE group. The most frequent fracture pattern was Neer type II found in 43 patients (64%). The overall lateral fragment size was 39 mm (± 12, range 14–83). There was a significant difference in bicortical lateral fragment size, 15 mm (± 4, range 6–21) in the CPH group compared to 20 mm (± 8, range 8–43) in the SCPLE group (p ≤ 0.001). The mean time from injury to surgery was 6.9 days (± 3.6, range 0–14). The mean follow-up was 37.5 months (± 17.9, range 12–76).

There was no significant difference in functional outcome, as shown in Table 2. The median QuickDASH score in the CHP group was 0.00 (IQR; 0.0–0.0), as opposed to 0.00 (IQR; 0.0–4.5) in the SCPLE group (p = 0.073). There were 15 patients (79%) with a QuickDASH score of 0 in the CHP group (range 0–21) compared to 25 patients (52%) in the SCPLE group (range 0–23). The median NRS pain score at rest was 0.00 (IQR; 0.0–0.0) in the CHP group and 0.00 (IQR; 0.0–0.0) in the SCPLE group (p = 0.373). There were 16 patients (84%) with a NRS pain score at rest of 0 in the CHP group (range 0–6) compared to 44 patients (94%) in the SCPLE group (range 0–3). In the CHP group, the median NRS pain score during activity was 0.00 (IQR; 0.0–1.0) compared to 0.00 (IQR; 0.0–2.0) in the SCPLE group (p = 0.559). There were 14 patients (74%) with a NRS pain score during activity of 0 in the CHP group (range 0–8) compared to 30 patients (63%) in the SCPLE group (range 0–7).

In both treatment groups, there was no significant difference in median QuickDASH score or other functional outcome scores between the Neer type II and type V fractures (Table 3). The median QuickDASH score in the Neer type II group following CHP fixation was 0.00 (IQR; 0.0–2.3), as opposed to 0.00 (IQR; 0.0–0.6) in the Neer type V group (p = 0.623). In the SCPLE group, the median QuickDASH score in the Neer type II group was 0.00 (IQR; 0.0–5.1), as opposed to 2.30 (IQR; 0.0–4.5) in the Neer type V group (p = 0.764).

Implant removal rates and indications are presented in Table 4. CHP fixation was associated with a significant higher removal rate. CHP removal was, according to protocol, performed in all 19 patients (100%) compared to 20 patients (42%) in the SCPLE group (relative risk 2.40; 95% CI 1.72–3.35; p ≤ 0.001). The mean time to removal was 4.3 months (± 2.2, range 2–10) and 13.6 months (± 11.5, range 5–50) in the CHP and SCPLE groups, respectively (mean difference − 9.287; 95% CI − 14.757 to 3.817; p = 0.002). In the CHP group, three patients (16%) reported removal without irritation and 16 patients (84%) reported removal due to irritation. There were no cases of implant removal-related complications. In the SCPLE group, 28 patients (58%) did not have the implant removed and 12 patients (43%) reported not to experience irritation.

Complications were reported in two patients (11%) in the CHP group compared to four patients (8%) in the SCPLE group (relative risk 1.26; 95% CI 0.25–6.33; p = 0.777) (Table 2). Complications in the CHP group consisted of one case of implant failure due to implant displacement and one case of non-union. Complications in the SCPLE group included three cases of implant failure and one case of non-union. The implant failures in SCPLE group consisted of two implant displacements and one case of screw breakage. No cases of infection or mal-union were observed. In total, there were three patients that needed revision surgery. In the CHP group, one patient received a lateral clavicle resection due to non-union. Two revision surgeries were performed in the SCPLE group, one due to severe implant displacement and one case of non-union. The SCPLE implant displacement was treated by repeat SCPLE fixation. The non-union was treated with temporary K-wires fixation for 9.5 months.