Midterm outcomes of 146 EVOLVE Proline modular radial head prostheses: a systematic review

Abstract

Introduction

The EVOLVE implant (Wright Medical Technology, Arlington, TN, USA) is a modular loose-fitting radial head prosthesis. The primary objective was to synthesize all available literature investigating the midterm clinical outcomes of the EVOLVE implant.

Materials and methods

An electronic literature search in Pubmed/Medline, Scopus, EMBASE, and Cochrane library was performed querying for studies published in 2000–2017. Articles describing clinical and radiographical outcomes as well as reoperation were included. Outcomes of interest included range of motion, Mayo Elbow Performance Score, Disabilities of the Arm Shoulder and Hand, radiographic outcome, and reason for reoperation.

Results

A total of five articles consisting of 146 patients with EVOLVE implants were included. Mean patient age was 57.4 years (range 22–84), and 43.8% were males (n = 64). Mean follow-up was 4.8 years (range 1–14). Mean Mayo Elbow Performance Score and Disabilities of the Arm Shoulder and Hand score were 87.6 (range 30–100) and 18.9 (range 0–82), respectively. Midterm clinical results were good or excellent (Mayo Elbow Performance Score > 74) in 94 patients. Reoperation was observed in 12 patients, with implant revision required in 2 patients. The primary reason for reoperation was persistent stiffness (n = 9).

Conclusion

Midterm outcomes of EVOLVE radial head prosthesis are satisfactory, and associated complication rates are low. Loose-fit implant method appears to be a reliable approach to avoid failure of radial head prosthesis by painful loosening.

Introduction

Radial head (RH) fractures account for a third of all fractures about the elbow.¹ The optimal treatment of Mason III RH fractures is controversial.^2–4 Radial head arthroplasty (RHA) and simple excision of the RH are alternatives in cases of fracture.^2–8 Excision provides satisfactory long-term results^9–11; however, it can also lead to valgus instability of the elbow, proximal migration of the radius, secondary ulnocarpal impaction, alteration in elbow kinematics, and/or accelerated ulnohumeral arthrosis.^6–8

Reconstruction of the lateral column of the elbow via RHA implants allows for maintenance of the native axis of rotation of the elbow and reduces the risk of both elbow and wrist arthrosis.^4–6 According to recent meta-analysis,¹² the failure rate of radial head arthroplasties varies between 0% and 45%; painful loosening is the primary reason for failure of RHA. Midterm survivorship of tight-fitting radial head prostheses (RHPs) is inferior to that of loose-fitting prostheses according to recent literature.^4,13–21

The EVOLVE Proline implant is a modular, unipolar, loose-fitting radial head implant system (Wright Medical Technology, Arlington, TN, USA) (Figure 1). Recent studies demonstrate that functional outcomes with the EVOLVE Proline implant are promising;^{10,12,14,22–26} the sample sizes in the aforementioned studies, however, are small. The primary objective of this study was to evaluate midterm clinical results of loose-fitting EVOLVE Proline RHAs. The primary hypothesis was that clinical results at midterm follow-up of loose-fitting EVOLVE Proline RHAs would be satisfactory. The secondary objective was to identify reasons for failure of these prostheses. The relevant hypotheses are that persistent stiffness would be a primary reason for failure among loosening prostheses.

Figure 1.

Image depicting the EVOLVE Proline radial head prosthesis with a modular unipolar design.

Materials and methods

Literature search strategy

An electronic literature search was carried out using PubMed/Medline, EMBASE, Scopus, and Cochrane library using Medical Subject Headings. The search was limited to English language literature. The following terms combined with « AND » and « OR »: « radial head »; « arthroplasty »; « prosthesis »; « radial head prosthesis »; and « radial head arthroplasty ». The reference list of each article was also searched in order to identify additional articles pertinent to our research criteria.

Selection criteria

The following inclusion criteria were established: (1) studies published between January 2000 and March 2017 and (2) describing clinical and radiographic results of RHA with EVOLVE Proline implants at a minimum of 1-year follow-up, in addition to reasons for failure. Exclusion criteria included: biomechanical studies, literature reviews and meta-analyses, case reports, reviews, comparative studies among radial head arthroplasties, and other treatment options, in series including implants other than the EVOLVE Proline.

Data extraction and critical appraisal

Data were extracted from any scripts, tables, and available figures. Two investigators (PL and MR) independently reviewed the full text of all included manuscripts. Age, sex, hand dominance, associated injuries, number of radial head arthroplasties, treatment timing, and survivorship following RHA were collected from all articles in order to allow for performance of a meta-analysis of all patients as a single group. Clinical results (range of motion of the elbow in flexion and extension as well as prono-supination, Mayo Elbow Performance Score (MEPS),²⁷ Disabilities of the Arm Shoulder and Hand (DASH)²⁸ score), radiographic results, and reasons for failure of the EVOLVE Proline prosthesis were also analyzed. When information on the variables of interest was missing, the corresponding authors were contacted.

Risk of bias assessment

The risk of bias was assessed utilizing criteria described by the Newcastle-Ottawa Quality Assessment Scale for observational studies.^29,30 Stars were placed according to the recommendations of the Newcastle-Ottawa Quality Assessment Scale: clear definition of the study population, clear definition of the results and how they were assessed, independent assessment of parameters, sufficient follow-up, lost to follow-up, and identification of prognostic and confounding factors.

Statistical analysis

Continuous variables are summarized using means and standard deviations, while categorically variables are represented by frequencies and proportions. Specific analyses accounting for confounding covariates were not possible due to a lack of available data. Statistical analysis was performed using open-source software (OpenMetaAnalyst).³¹ We defined p < 0.05 as statistically significant.

Results

Literature search results

Our initial literature search identified 1312 articles. After exclusion criteria were applied and duplicates were removed, 330 potentially eligible articles remained. Detailed review of the full texts of these articles led to inclusion of a total of five observational studies. Our search strategy is summarized in Figure 2.

Figure 2.

Summary of search strategy (Preferred Reporting Items for Systematic Reviews and Meta-Analyses—flow chart) for relevant studies on outcomes of EVOLVE Proline implants and their reasons for reoperation. ORIF: open reduction internal fixation.

Study and patient characteristics

One hundred forty-six patients were included in the final study. There were 64 males and 82 females with a mean age of 57.4 years (22–84 years). The dominant arm was involved in 50 cases. RHA was performed in an acute and delayed fashion in 143 and 3 cases, respectively. Among the acute RHAs, there were 51 isolated radial head fractures and 92 fractures associated with one or multiple additional injuries. Mean follow-up in the cohort was 4.8 years (range 1–14). Among acute injuries, elbow dislocation was the most common injury (n = 77; 54%), followed by coronoid fractures (n = 76; 53%), terrible triad (n = 61; 43%), and olecranon fractures (n = 16; 11%). There were only three cases of delayed repair reported in the literature, which were all attributed to development of pseudarthrosis. Study characteristics and patient demographics are reported in Table 1.

Table 1.

Patient demographics among article reporting outcomes of EVOLVE Proline implants.

	Marsh et al.14	Muhm et al.23	Chien et al.24	Doornberg et al.25	Grewal et al.26	Total
Country	Canada	Germany	Taiwan	USA	Canada
Year of publication	2016	2011	2010	2007	2006	2018
Study design	Retrospective	Retrospective	Retrospective	Retrospective	Prospective
Mean follow-up (years)	8.2 (5–14)	3.7 (1–7.2)	3.2 (1.7–5.8)	2.3 (2–4.6)	2 (2–4)	4.8 (1–14)
Patients characteristics	55	25	13	27	26	146
Men, n (%)	21 (38.2)	12 (48)	9 (69.2)	13 (48.1)	9 (34.6)	64 (43.8)
Age, mean	61	58.9	N/A	52	54	57.4
Dominant upper  limb up (years),  n (%)	26 (47.3)	14 (56)	N/A	7 (25.9)	11 (42.3)	58 (39.7)
Acute	55	25	10	27	26	143
Elbow dislocation	22	17	0	16	22	77
Coronoid fractures	32	14	4	13	13	76
Terrible triad	22	14	0	13	12	61
Olecranon fractures	3	0	1	10	2	16
Delayed	0	0	3	0	0	3
Pseudarthrosis	0	0	3	0	0	3

N/A: not available.

Clinical and radiographic outcomes

Clinical outcomes including range of motion of the elbow, an MEPS, and DASH score are presented in Table 2. Mean MEPS and DASH score were 87.6 (range 30–100) and 18.9 (range 0–82), respectively. MEPS were excellent or good among 92/113 acute RHAs (81%) and 2/3 delayed RHAs (67%). Given that the overwhelming majority of patients underwent arthroplasty for acute injury (143/146, 98%), we did not compare functional outcomes between acute and delayed arthroplasty cases.

Table 2.

Clinical outcome of radial head arthroplasty according to timing of treatment (acute or delayed).

	Acute RHA	Delayed RHA	Marsh et al.14	Muhm et al.23	Chien et al.24	Doornberg et al.25	Grewal et al.26	Overall
	N = 143	N = 3	N = 55	N = 25	N = 13	N = 27	N = 26	N = 146
Range of motion
Flexion–extension arc	118.7	101.6	126	114.3	120.4	111	113.2	118.3
Pronation	75.3	76.7	79	71.6	81.1	73	71	73.7
Supination	64.3	73.3	67	67.8	78.8	57	56.2	65.4
MEPS	n = 115	n = 3	n = 52	N/A	n = 13	n = 27	n = 26	n = 118
Mean	87.7	83.3	91		86.9	85	83.4	87.6
Categories
Excellent or   good (>74) (%)	92 (80)	2 (66.7)	45 (86.5)		11 (84.6)	22 (81.5)	16 (61.5)	94 (79.7)
Fair (60–73) (%)	15 (13)	1 (33)	5 (9.6)		2 (15.4)	3 (11.1)	6 (23.1)	16 (13.6)
Poor (<60) (%)	6 (5.2)	0	2 (3.6)		0	2 (7.4)	2 (7.7)	6 (5.1)
Dash score
Mean	20	37.7	14	26.1	N/A	17	24.4	18.9

RHA: radial head arthroplasty; MEPS: Mayo Elbow Performance Score; N/A: not available.

Periprosthetic osteolysis was identified radiographically in 76 cases (53%); localization of the osteolysis was not described among included studies. Radiographic results are reported in Table 3. A total of 12 (8%) reoperations after RHA were reported. There were nine reoperations with implant retention (6%) and two (1%) revisions. Persistent stiffness, corresponding to limitation in active and passive range of motion of the elbow after RHA,¹² was the only reason for revision in two cases (1%) and the primary reason for reoperation after RHA (n = 7; 5%). There was one case (1%) of persistent radial head subluxation following a terrible triad injury in which the coronoid was not fixed in a series described by Doornberg et al.²⁵ Ulnar nerve dysfunction (n = 2; 1%) one year after RHA was the second most frequently reported reason for reoperation with implant retention.²⁵

Table 3.

Radiographic outcomes of EVOLVE Proline implants in each study.

	Marsh et al.14	Muhm et al.23	Chien et al.24	Doornberg et al.25	Grewal et al.26	Total
	N = 55	N = 25	N = 13	N = 27	N = 26	N = 146
Periprosthetic osteolysis (%)	25 (45.4)	20 (80)	1 (7.7)	17 (63)	13 (50)	76 (52)
Heterotopic ossification (%)	20 (36.4)	19 (76)	0	13 (48.1)	6 (23.1)	58 (39.7)
Capitelar wear or osteopenia (%)	12 (21.8)	0	0	9 (33.3)	9 (34.6)	30 (20.5)
Overstuffing (equivalent oversizing) (%)	0	0	3 (23.1)	1 (3.7)	0	4 (2.7)
Ulnohumeral arthritis (%)	21 (38.2)	22 (88)	0	12 (44.4)	5 (19.2)	60 (41.1)

Assessment of bias

The risk of bias was low across all five studies (Supplemental Table 1).

Discussion

This study demonstrates that unipolar, loose-fitting EVOLVE Proline RHAs produce satisfactory midterm results and also corroborates the excellent results of RHA published in the recent literature.^{4,10–20,32,33} Preliminary functional outcomes of the EVOLVE Proline prosthesis have been promising; however, there is a paucity of data in the literature describing longer term outcomes with larger sample sizes.^14,23–26 According to a systematic review by Heijink et al.,¹⁰ tight-fitting RHAs led to superior functional outcomes at midterm follow-up when compared to loose-fitting implants. However, painful loosening, responsible for more than 25% of failures of RHAs, was a limiting factor for survivorship among press-fit implants.¹²

The rate of reoperation of EVOLVE Proline prostheses in the present era was low (n = 12; 8%). According to recent meta-analysis,¹² the mean rate of reoperation across all designs was 10%; painful loosening (n = 22; 27%) and persistent stiffness (n = 21; 26%) were the two primary reasons for failure of RHA. Persistent stiffness was the only reason for revision (n=2; 1%) and the primary reason for reoperation (n=7; 5%) among EVOLVE Proline prostheses. According to recent literature,^10,12 the rate of reoperation for persistent stiffness is significantly greater amount unipolar radial head prostheses. However, the causes of persistent stiffness after RHA are multifactorial, thereby introducing a confounding bias that could not be accounted for in the present study. Postoperative stiffness can be affected by the severity of the initial injury, heterotopic ossification, complex regional pain syndrome, capitellar degenerative lesions, or implant malpositioning.^34–39 We speculate that oscillation of the RH of unipolar implants against the capitellum, linked to micro-motion of loose-fitting stems, may lead to capitellar erosion (n = 30; 20%) and heterotopic ossification (n = 58, 40.6%), and may explain the elevated rate of persistent stiffness (n = 9, 6%) among EVOLVE Proline implants.^12,34–41 Stem micro-motion could also explain the elevated rate of periprosthetic osteolysis, while simultaneously limiting stress shielding due to the absence of a tight fit in the intramedullary canal.^33,40,41 The absence of available data regarding periprosthetic osteolysis did not allow for a detailed analysis of radiographic results of the EVOLVE Proline implant.

According to O’Driscoll et al.,^42–44 radiocapitellar stability of unipolar implants was significantly greater than that of bipolar implants. This finding was confirmed in multiple clinical studies, despite the fact that statistical significance could not be demonstrated.^42–44 According to Moon et al.,⁴³ radiocapitellar stability among unipolar implants can be explained by increased concavity compression among augmented implants. In cases where there is a ligamentous injury, unipolar prostheses are therefore the implant of choice.^43–45 The present study did not identify any reoperations for radiocapitellar instability. This confirms that micro-motion of the EVOLVE Proline RHP does not compromise radiocapitellar stability.

Biomechanical studies have demonstrated an increased risk of alteration in ulnohumeral kinematics as well as radiocapitellar stability linked to non-anatomic RHPs with an articular surface that is insufficiently deep and non-elliptical.^39,46–49 The implant design that most closely mimics proximal radius geometry is the ACUMED RHP (ACUMED, Hillsboro, OR, USA).⁵⁰ However, this design is a press-fit implant and requires optimal positioning and fixation in order to restore the anatomy of the proximal radius. It is therefore a technically challenging arthroplasty associated with a risk of painful loosening.^12,33,39,41 The modularity of the EVOLVE prosthesis, in addition to its smooth loose-fitting stem, allows for dynamic adaptation to the anatomy of the proximal radius all the while following biomechanical constraints of press-fit designs. According to recent meta-analysis,¹² painful loosening represents more than 25% of failures among RHAs, most notably among tight-fitting implants. Utilizing the definition of painful loosening described by O’Driscoll and Herald,^12,51 there were no cases reported among 143 patients in our series. Moghaddam et al.²² reported four failures (5.3%) of the EVOLVE Proline implant linked to painful loosening, wherein insidious proximal radius and forearm pain accentuated by radicapitellar motion (in the absence of capitellar lesions or implant malpositioning) was identified. Given the fact that the minimum follow-up among patients in this series was less than three years, the rate of painful loosening may be underestimated.¹² The EVOLVE Proline prosthesis only partially restores the anatomy of the RH; however, it allows for simple side implantation and a reduction in the risk of painful loosening.

Limitations of the present study are those inherent to single center, retrospective studies and include a potential lack of heterogeneity of the sample, loss to follow-up, and confounding bias. The 146 EVOLVE Proline prostheses were gathered from five single-center retrospective reports,^14,23–26 which may introduce a bias linked to the number of patients originating from the largest among the five series (Tables 1 to 3).¹⁴ The difference in sample size among RHAs performed acutely and in a delayed fashion did not allow for comparative analysis. Heterogeneity among associated injuries could not be taken into account in the analysis of clinical and radiographic outcomes. The minimum follow-up across all cohorts was less than three years¹²; we speculate that the rate of reoperation could therefore have been under estimated. However, a minimum follow-up of one year appears sufficient for the evaluation of functional result after RHA.⁵²

Conclusion

Midterm clinical results of the EVOLVE Proline prosthesis are satisfactory. The intentional loose-fitting design simplifies implantation and fixation while maintaining reliable midterm results. Additional long-term studies are needed to determine the risk of failure associated with loose-fitting implants.

Ethical Review and Patient Consent

Not required for this study.