- Email: [email protected]
Hemiarthroplasty Conversion: A Comparison to Primary and Revision Total Hip Arthroplasty
The Journal of Arthroplasty 34 (2019) 1168e1173
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Primary Arthroplasty
Hemiarthroplasty Conversion: A Comparison to Primary and Revision Total Hip Arthroplasty Nana O. Sarpong, MD, MBA *, Matthew J. Grosso, MD, Akshay Lakra, MD, Michael B. Held, MD, MBA, Carl L. Herndon, MD, H. John Cooper, MD Department of Orthopedic Surgery, Columbia University Irving Medical Center, New York, New York
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 December 2018 Received in revised form 13 February 2019 Accepted 20 February 2019 Available online 25 February 2019
Background: Arthroplasty is the standard of care for elderly patients with displaced femoral neck fractures, with viable options including hemiarthroplasty (HA) and total hip arthroplasty (THA). With time, HA may need to be converted to THA, but it is unclear whether this is more similar to primary or revision THA. We compare complication and revision rates between these groups within 90 days and 2 years postoperatively. Methods: We retrospectively reviewed 3 cohorts of patients treated at our institution: primary, conversion, and revision THA. Outcomes studied included intraoperative data, postoperative complications, and revision rates. We analyzed the groups using both parametric (analysis of variance test) and nonparametric (chi-squared test) statistics. Results: Operative time between primary THA (108.0 minutes), conversion HA (147.9 minutes), and revision THA (160.1 minutes) cohorts differed significantly (P ¼ .011). Estimated blood loss was also different between primary THA (386 mL), conversion HA (587 mL), and revision THA cohorts (529 mL) (P ¼ .011). At 2 years, major complication rates between primary THA (6.2%), conversion HA (11.7%), and revision THA (26.7%) cohorts also differed significantly (P ¼ .003), as was the revision rate in the primary THA (4.6%), conversion HA (10.0%), and revision THA (18.3%) cohorts (P ¼ .043). Conclusion: This is the first study to compare short-term and midterm complications between primary, conversion, and revision THA. We observed conversion HA had similar operative time and estimated blood loss to revision THA, which was significantly higher than primary THA. However, we found that conversion HA more closely resembled primary THA with respect to perioperative complications rates. © 2019 Elsevier Inc. All rights reserved.
Keywords: conversion THA revision THA hemiarthroplasty conversion complications revisions
By 2050, more than 6 million hip fractures are projected to occur annually, with 500,000 cases occurring in North America alone [1,2]. Hip fracture surgery has been associated with a 1-year mortality of 14% to 36% and represents a significant healthcare expenditure equating to over $80,000 per patient [2e4]. Hemiarthroplasty (HA) has historically been the most commonly used
This work was performed at the Department of Orthopedic Surgery, Columbia University Irving Medical Center, New York, NY. One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to https://doi.org/10.1016/j.arth.2019.02.043. * Reprint requests: Nana O. Sarpong, MD, MBA, Department of Orthopedic Surgery, Columbia University Irving Medical Center, 622 W 168th St, PH-11, New York, NY 10032. https://doi.org/10.1016/j.arth.2019.02.043 0883-5403/© 2019 Elsevier Inc. All rights reserved.
treatment for displaced femoral neck fractures in the elderly population, with the ultimate goal to restore patients to their preinjury ambulatory status [5]. Compared to total hip arthroplasty (THA), HA has been shown to have shorter operative times, less blood loss, lower dislocation rates, and lower initial costs as compared to primary THA for proximal femur fractures [6]. Although advancements in surgical technique and implant technology have led to improved implant survivorship, as the absolute number of arthroplasties continue to grow coupled with increasing life expectancies, a corresponding increase in the number of complications requiring conversion from a failed HA to THA will be expected [7e14], which may be associated with a second, often more-complex operation [15e23]. While there is limited literature on this topic, most studies examining conversion from HA to THA report high complication rates, with increased rates of periprosthetic dislocation, aseptic loosening, and periprosthetic fractures [24e27]. Overall, the evidence is limited with regard to
N.O. Sarpong et al. / The Journal of Arthroplasty 34 (2019) 1168e1173
whether conversion HA to THA is more similar to a primary THA for femoral neck fractures or to a revision THA procedure. To date, no study has directly compared outcomes between these cohorts. The present study compares postoperative complications and revision rates within 90 days and 2 years of the index procedure between 3 cohorts: primary THA for fractures, conversion HA to THA, and revision THA. We hypothesized that those who underwent primary THA for fractures would have fewer postoperative complications compared to those who underwent conversion HA to THA and revision THA. We predicted that revision THA would have a higher rate of complications compared to conversion HA to THA. Methods After institutional review board approval at our center was obtained, our institutional registry was used to identify 3 cohorts of patients: (1) patients who underwent primary THA for a hip fracture; (2) patients who underwent conversion from HA to a THA; and (3) patients who underwent revision THA after primary THA. All patients were treated at a single, tertiary care center. The first cohort (primary THA for fractures) was established by searching for Current Procedural Terminology (CPT) code 27130 for proximal femur fractures (International Classification of Diseases, 10th Revision S72.0, S72.1, S72.2) from January 1, 2007, to December 31, 2016, which yielded 65 patients. The second cohort (conversion HA to THA) was established by searching for CPT code 27132 from January 1, 2006, to December 31, 2016, which yielded 90 patients, but excluded cases of hip ORIF conversion to THA cases, leaving 60 cases available for the study. The third cohort (revision THA) was established by searching for CPT code 27138 from January 1, 2014, to December 31, 2016, which yielded 60 patients after excluding patients with prior hip resurfacing. Given that an all-inclusive revision THA cohort (polyethylene exchanges, both component, acetabular component, etc.) would have had significant intracohort variability, we chose to select only femoral component revisions to allow a better comparison to the conversion HA to THA cohort. Furthermore, the revision cohort was selected to match the reoperation indication in the conversion THA cohort (aseptic femoral loosening, periprosthetic fracture, periprosthetic dislocation). The primary THA, conversion HA to THA, and revision THA cohorts were matched for mean age, gender, body mass index, and American Society of Anesthesiologists score. There were no significant differences in the baseline demographic characteristics between the 3 cohorts (Table 1). The electronic medical records were then reviewed for each patient. Demographic information including age (at time of primary, conversion HA to THA, revision THA), gender, body mass index, American Society of Anesthesiologists score, and date of last follow-up was collected. Information regarding date of index and subsequent procedures, indication for the procedure, component fixation, and diameter of the femoral head component was also collected. For conversion and revision cases, we identified the reason for conversion or revision, and whether the femoral
Table 1 Patient Demographics. Group
Primary THA (N ¼ 65)
HA to THA (N ¼ 60)
Revision THA (N ¼ 60)
P Value
Age (y) Gender (%M/%F) BMI (kg/m2) ASA score
72.1 (35-92) 27.3/72.3 24.8 (17.4-39.5) 2.5
74.5 (31-94) 34.9/65.1 25.8 (18.2-36.6) 2.4
69.6 (44-94) 36.7/63.3 26.0 (17.7-40.7) 2.5
.097 .522 .231 .783
ASA, American Society of Anesthesiologists; BMI, body mass index; F, female; HA, hemiarthroplasty; M, male; THA, total hip arthroplasty.
1169
component was revised in the conversion case. Inpatient progress notes, operative reports, discharge summaries, and outpatient visit notes were reviewed in each cohort. Primary outcomes evaluated included postoperative major and minor complications and the need for subsequent reoperation or revision surgery within 90 days and 2 years of the primary, conversion, and revision THA. Major postoperative complications included periprosthetic fracture, prosthetic joint infection (PJI), periprosthetic dislocation, and aseptic femoral or acetabular loosening. Minor postoperative complications included superficial wound infection and hematoma. Secondary outcomes evaluated included operative time and estimated blood loss (EBL). A preliminary power analysis was performed with 80% power and demonstrated that 55 patients in each cohort would be needed for statistical significance, which was corroborated with a post hoc power analysis which revealed >85% power. Descriptive statistics were used for all data, using parametric (analysis of variance test) and nonparametric (chi-squared test) statistics to compare data between the 3 cohorts depending on the variable type. When outcomes exhibited a statistically significant difference among cohorts, we used the Student t-test to identify which 2 cohorts were different. A P value of .05 or less was considered statistically significant. Results Primary THA The primary THA cohort included 65 patients with an average follow-up of 1.95 years. In the primary THA cohort, the indication for the primary THA was for a femoral head/neck fracture in 61 (93.9%) and intertrochanteric fracture in 4 (6.1%) patients. Uncemented femoral fixation was used in 61 (93.9%) and uncemented acetabular fixation was used in 65 (100%) cases. The average femoral head diameter used was 33.5 mm (range, 28-40 mm). Acetabular supplemental fixation with screws was used in 37 (56.9%) cases. When used for supplemental acetabular fixation, the average number of screws used was 0.8. Conversion HA to THA The conversion HA to THA cohort included 60 patients with an average follow-up of 2.8 years. In the conversion HA to THA cohort, the indication for the index HA was for femoral head/neck fracture in 58 (96.7%) cases and intertrochanteric fractures in 2 (3.3%). The indications for conversion from HA to THA were for aseptic femoral loosening in 27 patients (45.0%), prosthetic arthritis in 20 patients (33.3%), periprosthetic fracture in 7 patients (11.7%), periprosthetic dislocation in 5 patients (8.3%), and leg length discrepancy in 1 patient (1.7%) (Figure 1). The average time from index hip HA to conversion THA was 7.4 years (range, 0.06-25.0 years). The average femoral head diameter used was 34.9 mm (range, 28-40 mm). During conversion surgery, the femoral component was revised in 75.0% and retained in 25.0% of cases. A hemispherical cementless acetabular component was added in 100% of cases. When used for supplemental acetabular fixation, the average number of screws was 1.2. Revision THA The revision THA cohort included 60 patients, with an average follow-up of 2.05 years. The indication for the index THA was for osteoarthritis in 45 patients (75.0%) and femoral head/neck fracture in 15 (25.0%). The indications for revision THA were for aseptic femoral loosening in 24 patients (40.0%), periprosthetic dislocation in 16 patients (26.7%), periprosthetic femoral fracture in 12 patients
1170
N.O. Sarpong et al. / The Journal of Arthroplasty 34 (2019) 1168e1173
Indications for Conversion THA Leg length discrepancy
1.67%
Dislocation Periprosthetic fracture
8.33% 11.67%
Prosthetic arthritis
33.33%
Aseptic loosening 0.0%
45.00% 5.0%
10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% 50.0% Percentage
Fig. 1. Indications for conversion HA to THA. HA, hemiarthroplasty; THA, total hip arthroplasty; PJI, prosthetic joint infection.
(20.0%), polyethylene wear in 3 patients (5.0%), leg length discrepancy in 3 patients (5.0%), and PJI treated in a single stage in 2 patients (3.3%). The average time from index THA to revision THA was 8.6 years (range, 0.01-18.5 years). As a result of our inclusion criteria for revision THA, the femoral component was revised in 100% of cases, to more closely match the conversion cohort. The average femoral head diameter used was 33.8 mm (range, 22-40 mm). Outcomes of Primary THA vs Conversion THA vs Revision THA The operative time between primary THA (108.0 minutes), conversion HA to THA (147.9 minutes), and revision THA (160.1 minutes) cohorts differed significantly (P ¼ .011; Table 2). When compared individually (Table 3), the operative time for primary THA was significantly less than conversion HA to THA (P ¼ .001) and revision THA (P ¼ .004). There was no significant difference in operative time between the conversion HA to THA and revision THA cohorts (P ¼ .489). Table 2 Comparison of Intraoperative and Postoperative Outcomes Within 90 Days and 2 Years Among Groups. Study Variable
Primary THA HA to THA
Operative time (min) EBL (mL) Femoral head diameter (mm, range) 90-d Total complications Major Minor 2-y Total complications Major complications (2 y) Dislocation PJI Periprosthetic fracture Aseptic loosening Minor complications Superficial infection Hematoma 90-d Revision rate 2-y Revision rate
108.0 385.5 33.5 (28-40)
147.9 160.1 586.9 529.2 34.9 (28-40) 33.8 (22-40)
Revision THA P Value .011 .011 .126
1 1 0 4 4 0 0 3 1
(1.5%) (1.5%) (0.0%) (6.2%) (6.2%) (0.0%) (0.0%) (4.6%) (1.5%)
2 2 0 7 6 4 1 0 1
(3.3%) (3.3%) (0.0%) (11.7%) (10%) (6.7%) (1.7%) (0.0%) (1.7%)
3 (5.0%) 2 (3.3%) 1 (1.7%) 16 (26.7%) 14 (23.3%) 6 (10.0%) 4 (6.7%) 2 (3.3%) 2 (3.3%)
.546 .785 .342 .003 .010 .041 .046 .248 .735
0 0 0 3
(0.0%) (0.0%) (0.0%) (4.6%)
0 1 2 6
(0.0%) (1.7%) (3.3%) (10.0%)
1 (1.7%) 1 (1.7%) 2 (3.3%) 11 (18.3%)
.342 .369 .339 .043
EBL, estimated blood loss; HA, hemiarthroplasty; PJI, prosthetic joint infection; THA, total hip arthroplasty. Bolded values represents the significance, P < .05.
The EBL also differed significantly among the primary THA (385.5 mL), conversion HA to THA (586.9 mL), and revision THA cohorts (529.2 mL; P ¼ .011; Table 2). When compared individually (Table 3), the EBL for the primary THA was significantly less than conversion HA to THA (P ¼ .008) and revision THA (P ¼ .014). There was no significant difference in EBL between the conversion HA to THA and revision THA cohorts (P ¼ .457). The 90-day total complication rate between the primary THA (1.5%, 1 major complication), conversion HA to THA (3.3%, 2 major complications), and revision THA cohorts (5.0%, 2 major and 1 minor complication) was not statistically different (P ¼ .546) (Table 2). The 2-year total complication rate between primary THA (6.2%, 4 complications), conversion HA to THA (11.7%, 7 complications), and revision THA (26.7%, 16 complications) differed significantly among the 3 cohorts (P ¼ .003; Table 2). At this time point, the complication rate between the primary THA and conversion HA to THA cohorts was not significantly different (P ¼ .507). However, the overall complication was significantly different between the primary THA and revision THA cohorts (P ¼ .0017) as well as between the conversion HA to THA and revision THA cohorts (P ¼ .036; Table 3). When overall complications within 2 years postoperatively were further stratified into major and minor complications, the major complication rate between primary THA (6.2%, 4 major complications), conversion HA to THA (10%, 6 major complications), and revision THA (23.3%, 14 major complications) differed significantly among the 3 cohorts (P ¼ .003; Table 2). The rate of major complications differed significantly between the primary and revision THA cohorts (P ¼ .0062), as well as between the conversion HA to THA and revision THA cohorts (P ¼ .040). There was no difference in the rate of major complications between the primary THA and conversion HA to THA cohorts (P ¼ .444; Table 3). When specific major complications at 2 years postoperatively were further stratified, significant differences were seen in 2 complication measures: periprosthetic dislocation and PJI. The periprosthetic dislocation rate in primary THA, conversion, and revision THA was 0.0% (0 patients), 6.7% (4 patients), and 10.0% (6 patients), respectively (P ¼ .041; Table 2). The periprosthetic dislocation rate in primary THA was significantly less compared to the rates in conversion HA (P ¼ .043) and revision THA (P ¼ .0089; Table 3). There was no significant difference in periprosthetic dislocation rates between the conversion HA to THA and revision
N.O. Sarpong et al. / The Journal of Arthroplasty 34 (2019) 1168e1173
1171
Table 3 Individual Comparison of Intraoperative and Postoperative Outcomes of Primary, Conversion HA to THA, and Revision THA When Significantly Different Among the 3 Cohorts. Outcomes
Primary vs Conversion THA
P Value
Primary vs Revision THA
Operative time (min) EBL (mL) Overall complications Major complications Dislocations PJI Revision
108.0 385.5 6.2% 6.2% 0.0% 0.0% 4.6%
.001 .008 .507 .444 .043 .307 .255
108.0 385.5 6.15% 6.2% 0.0% 0.0% 4.6%
147.9 586.9 11.7% 10.0% 6.7% 1.7% 10.0%
160.1 529.2 26.7% 20.0% 10.0% 6.7% 18.3%
P Value
Conversion vs Revision THA
.0045 .014 .0017 .0062 .0089 .0343 .0151
147.9 586.9 11.7% 10.0% 6.7% 1.7% 10.0%
P Value 160.1 529.2 26.7% 20.0% 10.0% 6.7% 18.3%
.489 .457 .036 .040 .508 .170 .190
EBL, estimated blood loss; HA, hemiarthroplasty; PJI, prosthetic joint infection; THA, total hip arthroplasty. Bolded values represents the significance, P < .05.
THA cohorts (P ¼ .118; Table 3). The PJI rate in primary, conversion HA to THA, and revision THA was 0.0% (0 patients), 1.7% (1 patient), and 6.7% (4 patients), respectively (P ¼ .046; Table 2). The PJI rate in primary THA and conversion HA to THA was not significantly different (P ¼ .307). The PJI rate in conversion HA to THA and revision THA cohorts was also not significantly different (P ¼ .170). The PJI rate was significantly higher in revision THA compared to primary THA (P ¼ .0343; Table 3). The rates of other major complications (periprosthetic fracture, aseptic loosening) and minor complications (hematoma, superficial wound infection) were not statistically different between the 3 cohorts (Table 2). Within 90 days postoperatively, the rate of revision surgery for the primary THA (0.0%), conversion HA to THA (3.3%, 2 revisions), and revision THA (3.3%, 2 revisions) cohorts was not significantly different (P ¼ .339; Table 2). However, at 2 years postoperatively, the rate of revision surgery for the primary THA (4.6%, 3 revisions), conversion HA to THA (10.0%, 6 revisions), and revision THA (18.3%, 11 revisions) cohorts differed significantly (P ¼ .043; Table 2). During this time period, the revision rate for the conversion HA to THA cohort was not significantly different from the revision rate for the primary THA cohort (P ¼ .255) and revision THA cohort (P ¼ .190). However, the re-revision rate in the revision THA cohort was significantly higher than the revision rate in primary THA (P ¼ .0151; Table 3). Discussion HA remains a mainstay treatment for displaced femoral neck fractures in the elderly population. While there are continued advancements in surgical technique and implant design, risk of conversion from an HA to a THA remains a significant concern. Our study documents a wide variety of reasons for possible failure of the original HA and subsequent conversion HA to THA. Despite an increasing prevalence of this procedure with increasing number of hip fractures and increased patient longevity, no studies have examined whether this procedure is more comparable to primary or revision THA. In this study, we observed that conversion HA to THA had similar operative time and EBL to revision THA, both significantly higher than primary THA. However, we found that conversion HA to THA more closely resembled primary THA with respect to perioperative complications rates within 2 years postoperatively. As for the rate of revision surgery in the conversion HA to THA cohort, we found it to be similar to both primary and revision THA. We did not observe any significant differences in the rates of major and minor complications, or revision surgery in the short term within 90 days postoperatively. Like other studies, periprosthetic dislocations accounted for the majority of major complications [28e30], and thus we questioned whether there was a correlation with the size of the femoral head component and dislocation rates. There was no statistically significant difference in the size
of the femoral head components, with mean head size above 32 mm in all 3 cohorts (Table 2). Our results highlight that femoral head size alone does not contribute to the increased rate of instability in the conversion HA to THA and revision THA cohorts. Furthermore, a subanalysis revealed that in our series, a dislocation event as an index indication for conversion and revision THA did not directly correlate with postoperative instability in the respective cohorts. A plausible theory for recurrent instability is the additional superior anterior capsular release in inserting the hemispherical acetabular cup during conversion HA to THA [31] as well as a higher rate of periprosthetic dislocation with utilization of the posterior approach during conversion and revision THA. We are not the first study to examine outcomes following conversion HA to THA surgery. In 2002, Sierra and Cabanela [24] identified 132 HA to THA conversion surgeries. They reported a high rate of major perioperative complications (45%), including a 10% rate of dislocations, and a 9% rate of periprosthetic femoral fracture, all higher than the rates reported in this study. In a prospective analysis, Pankaj et al [26] reported on 44 cases of conversion of HA to THA. Similar to our study, aseptic loosening and prosthetic arthritis were the primary causes of HA failure and conversion HA to THA. Once again, they reported a higher rate of perioperative complications compared to our study (23% vs 11.7%), primarily due to a high rate of periprosthetic fractures (20%), which is a more technically difficult operation than revision surgery for aseptic femoral loosening. Newer surgical techniques or implant designs may also have led to a decrease in perioperative complication rates, which may account for the lower rates of perioperative complications seen in our study. Most recently, Taheriazam and Saeidinia [32] evaluated outcomes in 138 patients who underwent HA to THA. Once again, stem loosening and prosthetic arthritis were the primary causes for failed HA. Similar to our study, they report relatively low rates of major perioperative complications (9%), with no periprosthetic fractures. Given the lack of literature and to gain a better understanding of conversion HA to THA, we chose to compare this procedure to primary THA and revision THA. By comparing these 3 cohorts, we aimed at better understanding the perioperative characteristics of conversion HA to THA, allowing surgeons to more easily correlate the risks associated with this procedure. There are no clinical cohort studies to date that compared the outcomes of conversion HA to THA, primary THA, and revision THA. However, based on a similar premise, using the American College of Surgeons National Surgical Quality Improvement Project database, Schwarzkopf and Baghoolizadeh [33] compared outcomes between primary THA, conversion THA, and revision THA. This comparison was based on CPT codes, and, therefore, the conversion THA group was a combination of internal fixation to THA and HA to THA. Their study reported a large number of differences between conversion and
1172
N.O. Sarpong et al. / The Journal of Arthroplasty 34 (2019) 1168e1173
primary THA and found that conversion THA was most similar to revision THA. Our study was unique, in that we chose to isolate conversion HA to THA, a very different procedure compared to internal fixation conversion. Similar to their study, we report significant differences between a primary THA and a conversion HA to THA. Although a comparison between studies is difficult because of different outcome measures and statistical methods, overall, our study found more differences between conversion HA to THA and revision THA, particularly for orthopedic-specific complications at time points within and beyond 30 days postoperatively, which is not provided by the National Surgical Quality Improvement Project database. In other words, based on perioperative complications, conversion HA to THA falls more closely toward primary THA than revision THA, compared to the Schwarzkopf and Baghoolizadeh’s [33] conclusion that conversion THA better resembles revision THA. One potential hypothesis for this difference is that conversion of an HA to THA may be a less complex procedure than conversion of an ORIF (such as cephalomedullary nail). This study has a number of limitations. First, study size for our patient cohorts was relatively small, compared to larger registry studies. Unfortunately, conversion HA to THA does not have a specific CPT code, so identifying this patient population in registrybased studies is difficult. In addition, follow-up in this study was limited to the midterm, with a mean follow-up of 1.95, 2.80, 2.05 years for the primary, conversion HA to THA, and revision cohorts, respectively. This limitation restricted our ability to examine longterm survival rates for the 3 cohorts. Rates of revision surgery are important outcome measures, and our relatively short follow-up period restricted our ability to compare these rates among groups more long term. However, our follow-up period is greater than that in previous reports [33], and we were able to stratify by short term (90 days) and midterm (2 years). Finally, there are a number of additional cohort comparisons that may be relevant. For example, we chose to compare conversion HA to THA to primary THA and to revision THA of only the femoral component. Additional possible comparisons include conversion HA to THA with or without femoral component exchange to the other 2 cohorts, conversion HA to THA to revision both component THA, and other variations. We felt that the 3 cohort comparisons we chose was an appropriate balance of study feasibility and clinical relevance.
[3]
[4] [5]
[6]
[7] [8]
[9]
[10] [11]
[12]
[13]
[14] [15]
[16]
[17]
[18]
Conclusion
[19]
Conversion surgery to THA is a solution for failed HAs following femoral neck fractures. Our understanding of the outcomes and complications following this unique surgical procedure is limited. This is the first clinical study to compare complication and revision rates between conversion HA to THA, primary THA, and revision THA. In this study, conversion HA to THA more closely resembled primary THA than revision THA in overall complications within 2 years postoperatively. For operative time, blood loss, and periprosthetic dislocations, conversion HA to THA more closely resembled revision THA. By comparing conversion HA to THA to the more commonly performed procedures of primary THA and revision THA, the reported outcomes in this study can help surgeons appropriately stratify the risks involved in this surgery for this atrisk patient population.
[20]
[21]
[22]
[23]
[24] [25]
References [1] Bhandari M, Devereaux PJ, Tornetta P, Swiontkowski MF, Berry DJ, Haidukewych G, et al. Operative management of displaced femoral neck fractures in elderly patients. An international survey. J Bone Joint Surg Am 2005;87:2122e30. https://doi.org/10.2106/JBJS.E.00535. [2] Evaniew N, Madden K, Bhandari M. Cochrane in CORR®: arthroplasties (with and without bone cement) for proximal femoral fractures in adults. Clin
[26]
[27]
Orthop Relat Res 2014;472:1367e72. https://doi.org/10.1007/s11999-0133328-5. Ahn J, Man L-X, Park S, Sodl JF, Esterhai JL. Systematic review of cemented and uncemented hemiarthroplasty outcomes for femoral neck fractures. Clin Orthop Relat Res 2008;466:2513e8. https://doi.org/10.1007/s11999-0080368-3. Braithwaite RS, Col NF, Wong JB. Estimating hip fracture morbidity, mortality and costs. J Am Geriatr Soc 2003;51:364e70. Grosso MG, Danoff JR, Padgett DE, Iorio R, Macaulay WB. The cemented unipolar prosthesis for the management of displaced femoral neck fractures in the dependent osteopenic elderly. J Arthroplasty 2016;31:1040e6. https:// doi.org/10.1016/j.arth.2015.11.029. Lee BP, Berry DJ, Harmsen WS, Sim FH. Total hip arthroplasty for the treatment of an acute fracture of the femoral neck: long-term results. J Bone Joint Surg Am 1998;80:70e5. Bourne RB, Maloney WJ, Wright JG. An AOA critical issue. The outcome of the outcomes movement. J Bone Joint Surg Am 2004;86-A:633e40. Ong KL, Lau E, Suggs J, Kurtz SM, Manley MT. Risk of subsequent revision after primary and revision total joint arthroplasty. Clin Orthop Relat Res 2010;468: 3070e6. https://doi.org/10.1007/s11999-010-1399-0. Makarewich CA, Anderson MB, Gililland JM, Pelt CE, Peters CL. Ten-year survivorship of primary total hip arthroplasty in patients 30 years of age or younger. Bone Joint J 2018;100-B:867e74. https://doi.org/10.1302/0301620X.100B7.BJJ-2017-1603.R1. Pakos EE, Paschos NK, Xenakis TA. Long term outcomes of total hip arthroplasty in young patients under 30. Arch Bone Jt Surg 2014;2:157e62. Girard J, Bocquet D, Autissier G, Fouilleron N, Fron D, Migaud H. Metal-onmetal hip arthroplasty in patients thirty years of age or younger. J Bone Joint Surg Am 2010;92:2419e26. https://doi.org/10.2106/JBJS.I.01644. Chin G, Wright DJ, Snir N, Schwarzkopf R. Primary vs conversion total hip arthroplasty: a cost analysis. J Arthroplasty 2016;31:362e7. https://doi.org/ 10.1016/j.arth.2015.08.014. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 2007;89:780e5. https://doi.org/10.2106/JBJS.F.00222. Gullberg B, Johnell O, Kanis JA. World-wide projections for hip fracture. Osteoporos Int 1997;7:407e13. Burgers PT, Van Geene AR, Van den Bekerom MP, Van Lieshout EM, Blom B, Aleem IS, et al. Total hip arthroplasty versus hemiarthroplasty for displaced femoral neck fractures in the healthy elderly: a meta-analysis and systematic review of randomized trials. Int Orthop 2012;36:1549e60. https://doi.org/ 10.1007/s00264-012-1569-7. Keating JF, Grant A, Masson M, Scott NW, Forbes JF. Displaced intracapsular hip fractures in fit, older people: a randomised comparison of reduction and fixation, bipolar hemiarthroplasty and total hip arthroplasty. Health Technol Assess 2005;9:iiieiv. ixex, 1e65. van den Bekerom MP, Sierevelt IN, Bonke H, Raaymakers EL. The natural history of the hemiarthroplasty for displaced intracapsular femoral neck fractures. Acta Orthop 2013;84:555e60. https://doi.org/10.3109/ 17453674.2013.867763. Grosso MJ, Danoff JR, Murtaugh TS, Trofa DP, Sawires AN, Macaulay WB. Hemiarthroplasty for displaced femoral neck fractures in the elderly has a low conversion rate. J Arthroplasty 2017;32:150e4. https://doi.org/10.1016/ j.arth.2016.06.048. Wachtl SW, Jakob RP, Gautier E. Ten-year patient and prosthesis survival after unipolar hip hemiarthroplasty in female patients over 70 years old. J Arthroplasty 2003;18:587e91. Avery PP, Baker RP, Walton MJ, Rooker JC, Squires B, Gargan MF, et al. Total hip replacement and hemiarthroplasty in mobile, independent patients with a displaced intracapsular fracture of the femoral neck: a seven- to ten-year follow-up report of a prospective randomised controlled trial. J Bone Joint Surg Br 2011;93:1045e8. https://doi.org/10.1302/0301620X.93B8.27132. Alazzawi S, Sprenger De Rover WB, Brown J, Davis B. The conversion rate of bipolar hemiarthroplasty after a hip fracture to a total hip arthroplasty. Clin Orthop Surg 2012;4:117e20. https://doi.org/10.4055/cios.2012.4.2.117. Tanous T, Stephenson KW, Grecula MJ. Hip hemiarthroplasty after displaced femoral neck fracture: a survivorship analysis. Orthopedics 2010;33:385. https://doi.org/10.3928/01477447-20100429-08. Grosso MJ, Danoff JR, Thacher R, Murtaugh TS, Hickernell TR, Shah RP, et al. Risk factors for conversion surgery to total hip arthroplasty of a hemiarthroplasty performed for a femoral neck fracture. Hip Int 2018;28:168e72. https://doi.org/10.1177/1120700018768654. Sierra RJ, Cabanela ME. Conversion of failed hip hemiarthroplasties after femoral neck fractures. Clin Orthop Relat Res 2002;399:129e39. Figved W, Dybvik E, Frihagen F, Furnes O, Madsen JE, Havelin LI, et al. Conversion from failed hemiarthroplasty to total hip arthroplasty: a Norwegian Arthroplasty Register analysis of 595 hips with previous femoral neck fractures. Acta Orthop 2007;78:711e8. https://doi.org/10.1080/ 17453670710014473. Pankaj A, Malhotra R, Bhan S. Conversion of failed hemiarthroplasty to total hip arthroplasty: a short to mid-term follow-up study. Indian J Orthop 2008;42:294e300. https://doi.org/10.4103/0019-5413.41852. Fichman SG, M€ akinen TJ, Vincent A, Lozano B, Safir O, Kuzyk PRT. Complications following conversion of a hip hemiarthroplasty to a total hip
N.O. Sarpong et al. / The Journal of Arthroplasty 34 (2019) 1168e1173 arthroplasty. Int Orthop 2015;39:2335e9. https://doi.org/10.1007/s00264015-2827-2. [28] Sharkey PF, Rao R, Hozack WJ, Rothman RH, Carey C. Conversion of hemiarthroplasty to total hip arthroplasty: can groin pain be eliminated? J Arthroplasty 1998;13:627e30. [29] Sarmiento A, Gerard FM. Total hip arthroplasty for failed endoprostheses. Clin Orthop Relat Res 1978;137:112e7. [30] Sah AP, Estok DM. Dislocation rate after conversion from hip hemiarthroplasty to total hip arthroplasty. J Bone Joint Surg Am 2008;90:506e16. https:// doi.org/10.2106/JBJS.G.00479.
1173
[31] Park K-S, Chan C-K, Lee D-H, Yoon T-R. Midterm results of conversion from failed bipolar hemiarthroplasty to total hip arthroplasty. Indian J Orthop 2018;52:369e73. https://doi.org/10.4103/ortho.IJOrtho_ 494_16. [32] Taheriazam A, Saeidinia A. Conversion of failed hemiarthroplasty to total hip arthroplasty: a short-term follow-up study. Medicine (Baltimore) 2017;96: e8235. https://doi.org/10.1097/MD.0000000000008235. [33] Schwarzkopf R, Baghoolizadeh M. Conversion total hip arthroplasty: primary or revision total hip arthroplasty. World J Orthop 2015;6:750e3. https:// doi.org/10.5312/wjo.v6.i10.750.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Primary Arthroplasty
Hemiarthroplasty Conversion: A Comparison to Primary and Revision Total Hip Arthroplasty Nana O. Sarpong, MD, MBA *, Matthew J. Grosso, MD, Akshay Lakra, MD, Michael B. Held, MD, MBA, Carl L. Herndon, MD, H. John Cooper, MD Department of Orthopedic Surgery, Columbia University Irving Medical Center, New York, New York
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 December 2018 Received in revised form 13 February 2019 Accepted 20 February 2019 Available online 25 February 2019
Background: Arthroplasty is the standard of care for elderly patients with displaced femoral neck fractures, with viable options including hemiarthroplasty (HA) and total hip arthroplasty (THA). With time, HA may need to be converted to THA, but it is unclear whether this is more similar to primary or revision THA. We compare complication and revision rates between these groups within 90 days and 2 years postoperatively. Methods: We retrospectively reviewed 3 cohorts of patients treated at our institution: primary, conversion, and revision THA. Outcomes studied included intraoperative data, postoperative complications, and revision rates. We analyzed the groups using both parametric (analysis of variance test) and nonparametric (chi-squared test) statistics. Results: Operative time between primary THA (108.0 minutes), conversion HA (147.9 minutes), and revision THA (160.1 minutes) cohorts differed significantly (P ¼ .011). Estimated blood loss was also different between primary THA (386 mL), conversion HA (587 mL), and revision THA cohorts (529 mL) (P ¼ .011). At 2 years, major complication rates between primary THA (6.2%), conversion HA (11.7%), and revision THA (26.7%) cohorts also differed significantly (P ¼ .003), as was the revision rate in the primary THA (4.6%), conversion HA (10.0%), and revision THA (18.3%) cohorts (P ¼ .043). Conclusion: This is the first study to compare short-term and midterm complications between primary, conversion, and revision THA. We observed conversion HA had similar operative time and estimated blood loss to revision THA, which was significantly higher than primary THA. However, we found that conversion HA more closely resembled primary THA with respect to perioperative complications rates. © 2019 Elsevier Inc. All rights reserved.
Keywords: conversion THA revision THA hemiarthroplasty conversion complications revisions
By 2050, more than 6 million hip fractures are projected to occur annually, with 500,000 cases occurring in North America alone [1,2]. Hip fracture surgery has been associated with a 1-year mortality of 14% to 36% and represents a significant healthcare expenditure equating to over $80,000 per patient [2e4]. Hemiarthroplasty (HA) has historically been the most commonly used
This work was performed at the Department of Orthopedic Surgery, Columbia University Irving Medical Center, New York, NY. One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to https://doi.org/10.1016/j.arth.2019.02.043. * Reprint requests: Nana O. Sarpong, MD, MBA, Department of Orthopedic Surgery, Columbia University Irving Medical Center, 622 W 168th St, PH-11, New York, NY 10032. https://doi.org/10.1016/j.arth.2019.02.043 0883-5403/© 2019 Elsevier Inc. All rights reserved.
treatment for displaced femoral neck fractures in the elderly population, with the ultimate goal to restore patients to their preinjury ambulatory status [5]. Compared to total hip arthroplasty (THA), HA has been shown to have shorter operative times, less blood loss, lower dislocation rates, and lower initial costs as compared to primary THA for proximal femur fractures [6]. Although advancements in surgical technique and implant technology have led to improved implant survivorship, as the absolute number of arthroplasties continue to grow coupled with increasing life expectancies, a corresponding increase in the number of complications requiring conversion from a failed HA to THA will be expected [7e14], which may be associated with a second, often more-complex operation [15e23]. While there is limited literature on this topic, most studies examining conversion from HA to THA report high complication rates, with increased rates of periprosthetic dislocation, aseptic loosening, and periprosthetic fractures [24e27]. Overall, the evidence is limited with regard to
N.O. Sarpong et al. / The Journal of Arthroplasty 34 (2019) 1168e1173
whether conversion HA to THA is more similar to a primary THA for femoral neck fractures or to a revision THA procedure. To date, no study has directly compared outcomes between these cohorts. The present study compares postoperative complications and revision rates within 90 days and 2 years of the index procedure between 3 cohorts: primary THA for fractures, conversion HA to THA, and revision THA. We hypothesized that those who underwent primary THA for fractures would have fewer postoperative complications compared to those who underwent conversion HA to THA and revision THA. We predicted that revision THA would have a higher rate of complications compared to conversion HA to THA. Methods After institutional review board approval at our center was obtained, our institutional registry was used to identify 3 cohorts of patients: (1) patients who underwent primary THA for a hip fracture; (2) patients who underwent conversion from HA to a THA; and (3) patients who underwent revision THA after primary THA. All patients were treated at a single, tertiary care center. The first cohort (primary THA for fractures) was established by searching for Current Procedural Terminology (CPT) code 27130 for proximal femur fractures (International Classification of Diseases, 10th Revision S72.0, S72.1, S72.2) from January 1, 2007, to December 31, 2016, which yielded 65 patients. The second cohort (conversion HA to THA) was established by searching for CPT code 27132 from January 1, 2006, to December 31, 2016, which yielded 90 patients, but excluded cases of hip ORIF conversion to THA cases, leaving 60 cases available for the study. The third cohort (revision THA) was established by searching for CPT code 27138 from January 1, 2014, to December 31, 2016, which yielded 60 patients after excluding patients with prior hip resurfacing. Given that an all-inclusive revision THA cohort (polyethylene exchanges, both component, acetabular component, etc.) would have had significant intracohort variability, we chose to select only femoral component revisions to allow a better comparison to the conversion HA to THA cohort. Furthermore, the revision cohort was selected to match the reoperation indication in the conversion THA cohort (aseptic femoral loosening, periprosthetic fracture, periprosthetic dislocation). The primary THA, conversion HA to THA, and revision THA cohorts were matched for mean age, gender, body mass index, and American Society of Anesthesiologists score. There were no significant differences in the baseline demographic characteristics between the 3 cohorts (Table 1). The electronic medical records were then reviewed for each patient. Demographic information including age (at time of primary, conversion HA to THA, revision THA), gender, body mass index, American Society of Anesthesiologists score, and date of last follow-up was collected. Information regarding date of index and subsequent procedures, indication for the procedure, component fixation, and diameter of the femoral head component was also collected. For conversion and revision cases, we identified the reason for conversion or revision, and whether the femoral
Table 1 Patient Demographics. Group
Primary THA (N ¼ 65)
HA to THA (N ¼ 60)
Revision THA (N ¼ 60)
P Value
Age (y) Gender (%M/%F) BMI (kg/m2) ASA score
72.1 (35-92) 27.3/72.3 24.8 (17.4-39.5) 2.5
74.5 (31-94) 34.9/65.1 25.8 (18.2-36.6) 2.4
69.6 (44-94) 36.7/63.3 26.0 (17.7-40.7) 2.5
.097 .522 .231 .783
ASA, American Society of Anesthesiologists; BMI, body mass index; F, female; HA, hemiarthroplasty; M, male; THA, total hip arthroplasty.
1169
component was revised in the conversion case. Inpatient progress notes, operative reports, discharge summaries, and outpatient visit notes were reviewed in each cohort. Primary outcomes evaluated included postoperative major and minor complications and the need for subsequent reoperation or revision surgery within 90 days and 2 years of the primary, conversion, and revision THA. Major postoperative complications included periprosthetic fracture, prosthetic joint infection (PJI), periprosthetic dislocation, and aseptic femoral or acetabular loosening. Minor postoperative complications included superficial wound infection and hematoma. Secondary outcomes evaluated included operative time and estimated blood loss (EBL). A preliminary power analysis was performed with 80% power and demonstrated that 55 patients in each cohort would be needed for statistical significance, which was corroborated with a post hoc power analysis which revealed >85% power. Descriptive statistics were used for all data, using parametric (analysis of variance test) and nonparametric (chi-squared test) statistics to compare data between the 3 cohorts depending on the variable type. When outcomes exhibited a statistically significant difference among cohorts, we used the Student t-test to identify which 2 cohorts were different. A P value of .05 or less was considered statistically significant. Results Primary THA The primary THA cohort included 65 patients with an average follow-up of 1.95 years. In the primary THA cohort, the indication for the primary THA was for a femoral head/neck fracture in 61 (93.9%) and intertrochanteric fracture in 4 (6.1%) patients. Uncemented femoral fixation was used in 61 (93.9%) and uncemented acetabular fixation was used in 65 (100%) cases. The average femoral head diameter used was 33.5 mm (range, 28-40 mm). Acetabular supplemental fixation with screws was used in 37 (56.9%) cases. When used for supplemental acetabular fixation, the average number of screws used was 0.8. Conversion HA to THA The conversion HA to THA cohort included 60 patients with an average follow-up of 2.8 years. In the conversion HA to THA cohort, the indication for the index HA was for femoral head/neck fracture in 58 (96.7%) cases and intertrochanteric fractures in 2 (3.3%). The indications for conversion from HA to THA were for aseptic femoral loosening in 27 patients (45.0%), prosthetic arthritis in 20 patients (33.3%), periprosthetic fracture in 7 patients (11.7%), periprosthetic dislocation in 5 patients (8.3%), and leg length discrepancy in 1 patient (1.7%) (Figure 1). The average time from index hip HA to conversion THA was 7.4 years (range, 0.06-25.0 years). The average femoral head diameter used was 34.9 mm (range, 28-40 mm). During conversion surgery, the femoral component was revised in 75.0% and retained in 25.0% of cases. A hemispherical cementless acetabular component was added in 100% of cases. When used for supplemental acetabular fixation, the average number of screws was 1.2. Revision THA The revision THA cohort included 60 patients, with an average follow-up of 2.05 years. The indication for the index THA was for osteoarthritis in 45 patients (75.0%) and femoral head/neck fracture in 15 (25.0%). The indications for revision THA were for aseptic femoral loosening in 24 patients (40.0%), periprosthetic dislocation in 16 patients (26.7%), periprosthetic femoral fracture in 12 patients
1170
N.O. Sarpong et al. / The Journal of Arthroplasty 34 (2019) 1168e1173
Indications for Conversion THA Leg length discrepancy
1.67%
Dislocation Periprosthetic fracture
8.33% 11.67%
Prosthetic arthritis
33.33%
Aseptic loosening 0.0%
45.00% 5.0%
10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% 50.0% Percentage
Fig. 1. Indications for conversion HA to THA. HA, hemiarthroplasty; THA, total hip arthroplasty; PJI, prosthetic joint infection.
(20.0%), polyethylene wear in 3 patients (5.0%), leg length discrepancy in 3 patients (5.0%), and PJI treated in a single stage in 2 patients (3.3%). The average time from index THA to revision THA was 8.6 years (range, 0.01-18.5 years). As a result of our inclusion criteria for revision THA, the femoral component was revised in 100% of cases, to more closely match the conversion cohort. The average femoral head diameter used was 33.8 mm (range, 22-40 mm). Outcomes of Primary THA vs Conversion THA vs Revision THA The operative time between primary THA (108.0 minutes), conversion HA to THA (147.9 minutes), and revision THA (160.1 minutes) cohorts differed significantly (P ¼ .011; Table 2). When compared individually (Table 3), the operative time for primary THA was significantly less than conversion HA to THA (P ¼ .001) and revision THA (P ¼ .004). There was no significant difference in operative time between the conversion HA to THA and revision THA cohorts (P ¼ .489). Table 2 Comparison of Intraoperative and Postoperative Outcomes Within 90 Days and 2 Years Among Groups. Study Variable
Primary THA HA to THA
Operative time (min) EBL (mL) Femoral head diameter (mm, range) 90-d Total complications Major Minor 2-y Total complications Major complications (2 y) Dislocation PJI Periprosthetic fracture Aseptic loosening Minor complications Superficial infection Hematoma 90-d Revision rate 2-y Revision rate
108.0 385.5 33.5 (28-40)
147.9 160.1 586.9 529.2 34.9 (28-40) 33.8 (22-40)
Revision THA P Value .011 .011 .126
1 1 0 4 4 0 0 3 1
(1.5%) (1.5%) (0.0%) (6.2%) (6.2%) (0.0%) (0.0%) (4.6%) (1.5%)
2 2 0 7 6 4 1 0 1
(3.3%) (3.3%) (0.0%) (11.7%) (10%) (6.7%) (1.7%) (0.0%) (1.7%)
3 (5.0%) 2 (3.3%) 1 (1.7%) 16 (26.7%) 14 (23.3%) 6 (10.0%) 4 (6.7%) 2 (3.3%) 2 (3.3%)
.546 .785 .342 .003 .010 .041 .046 .248 .735
0 0 0 3
(0.0%) (0.0%) (0.0%) (4.6%)
0 1 2 6
(0.0%) (1.7%) (3.3%) (10.0%)
1 (1.7%) 1 (1.7%) 2 (3.3%) 11 (18.3%)
.342 .369 .339 .043
EBL, estimated blood loss; HA, hemiarthroplasty; PJI, prosthetic joint infection; THA, total hip arthroplasty. Bolded values represents the significance, P < .05.
The EBL also differed significantly among the primary THA (385.5 mL), conversion HA to THA (586.9 mL), and revision THA cohorts (529.2 mL; P ¼ .011; Table 2). When compared individually (Table 3), the EBL for the primary THA was significantly less than conversion HA to THA (P ¼ .008) and revision THA (P ¼ .014). There was no significant difference in EBL between the conversion HA to THA and revision THA cohorts (P ¼ .457). The 90-day total complication rate between the primary THA (1.5%, 1 major complication), conversion HA to THA (3.3%, 2 major complications), and revision THA cohorts (5.0%, 2 major and 1 minor complication) was not statistically different (P ¼ .546) (Table 2). The 2-year total complication rate between primary THA (6.2%, 4 complications), conversion HA to THA (11.7%, 7 complications), and revision THA (26.7%, 16 complications) differed significantly among the 3 cohorts (P ¼ .003; Table 2). At this time point, the complication rate between the primary THA and conversion HA to THA cohorts was not significantly different (P ¼ .507). However, the overall complication was significantly different between the primary THA and revision THA cohorts (P ¼ .0017) as well as between the conversion HA to THA and revision THA cohorts (P ¼ .036; Table 3). When overall complications within 2 years postoperatively were further stratified into major and minor complications, the major complication rate between primary THA (6.2%, 4 major complications), conversion HA to THA (10%, 6 major complications), and revision THA (23.3%, 14 major complications) differed significantly among the 3 cohorts (P ¼ .003; Table 2). The rate of major complications differed significantly between the primary and revision THA cohorts (P ¼ .0062), as well as between the conversion HA to THA and revision THA cohorts (P ¼ .040). There was no difference in the rate of major complications between the primary THA and conversion HA to THA cohorts (P ¼ .444; Table 3). When specific major complications at 2 years postoperatively were further stratified, significant differences were seen in 2 complication measures: periprosthetic dislocation and PJI. The periprosthetic dislocation rate in primary THA, conversion, and revision THA was 0.0% (0 patients), 6.7% (4 patients), and 10.0% (6 patients), respectively (P ¼ .041; Table 2). The periprosthetic dislocation rate in primary THA was significantly less compared to the rates in conversion HA (P ¼ .043) and revision THA (P ¼ .0089; Table 3). There was no significant difference in periprosthetic dislocation rates between the conversion HA to THA and revision
N.O. Sarpong et al. / The Journal of Arthroplasty 34 (2019) 1168e1173
1171
Table 3 Individual Comparison of Intraoperative and Postoperative Outcomes of Primary, Conversion HA to THA, and Revision THA When Significantly Different Among the 3 Cohorts. Outcomes
Primary vs Conversion THA
P Value
Primary vs Revision THA
Operative time (min) EBL (mL) Overall complications Major complications Dislocations PJI Revision
108.0 385.5 6.2% 6.2% 0.0% 0.0% 4.6%
.001 .008 .507 .444 .043 .307 .255
108.0 385.5 6.15% 6.2% 0.0% 0.0% 4.6%
147.9 586.9 11.7% 10.0% 6.7% 1.7% 10.0%
160.1 529.2 26.7% 20.0% 10.0% 6.7% 18.3%
P Value
Conversion vs Revision THA
.0045 .014 .0017 .0062 .0089 .0343 .0151
147.9 586.9 11.7% 10.0% 6.7% 1.7% 10.0%
P Value 160.1 529.2 26.7% 20.0% 10.0% 6.7% 18.3%
.489 .457 .036 .040 .508 .170 .190
EBL, estimated blood loss; HA, hemiarthroplasty; PJI, prosthetic joint infection; THA, total hip arthroplasty. Bolded values represents the significance, P < .05.
THA cohorts (P ¼ .118; Table 3). The PJI rate in primary, conversion HA to THA, and revision THA was 0.0% (0 patients), 1.7% (1 patient), and 6.7% (4 patients), respectively (P ¼ .046; Table 2). The PJI rate in primary THA and conversion HA to THA was not significantly different (P ¼ .307). The PJI rate in conversion HA to THA and revision THA cohorts was also not significantly different (P ¼ .170). The PJI rate was significantly higher in revision THA compared to primary THA (P ¼ .0343; Table 3). The rates of other major complications (periprosthetic fracture, aseptic loosening) and minor complications (hematoma, superficial wound infection) were not statistically different between the 3 cohorts (Table 2). Within 90 days postoperatively, the rate of revision surgery for the primary THA (0.0%), conversion HA to THA (3.3%, 2 revisions), and revision THA (3.3%, 2 revisions) cohorts was not significantly different (P ¼ .339; Table 2). However, at 2 years postoperatively, the rate of revision surgery for the primary THA (4.6%, 3 revisions), conversion HA to THA (10.0%, 6 revisions), and revision THA (18.3%, 11 revisions) cohorts differed significantly (P ¼ .043; Table 2). During this time period, the revision rate for the conversion HA to THA cohort was not significantly different from the revision rate for the primary THA cohort (P ¼ .255) and revision THA cohort (P ¼ .190). However, the re-revision rate in the revision THA cohort was significantly higher than the revision rate in primary THA (P ¼ .0151; Table 3). Discussion HA remains a mainstay treatment for displaced femoral neck fractures in the elderly population. While there are continued advancements in surgical technique and implant design, risk of conversion from an HA to a THA remains a significant concern. Our study documents a wide variety of reasons for possible failure of the original HA and subsequent conversion HA to THA. Despite an increasing prevalence of this procedure with increasing number of hip fractures and increased patient longevity, no studies have examined whether this procedure is more comparable to primary or revision THA. In this study, we observed that conversion HA to THA had similar operative time and EBL to revision THA, both significantly higher than primary THA. However, we found that conversion HA to THA more closely resembled primary THA with respect to perioperative complications rates within 2 years postoperatively. As for the rate of revision surgery in the conversion HA to THA cohort, we found it to be similar to both primary and revision THA. We did not observe any significant differences in the rates of major and minor complications, or revision surgery in the short term within 90 days postoperatively. Like other studies, periprosthetic dislocations accounted for the majority of major complications [28e30], and thus we questioned whether there was a correlation with the size of the femoral head component and dislocation rates. There was no statistically significant difference in the size
of the femoral head components, with mean head size above 32 mm in all 3 cohorts (Table 2). Our results highlight that femoral head size alone does not contribute to the increased rate of instability in the conversion HA to THA and revision THA cohorts. Furthermore, a subanalysis revealed that in our series, a dislocation event as an index indication for conversion and revision THA did not directly correlate with postoperative instability in the respective cohorts. A plausible theory for recurrent instability is the additional superior anterior capsular release in inserting the hemispherical acetabular cup during conversion HA to THA [31] as well as a higher rate of periprosthetic dislocation with utilization of the posterior approach during conversion and revision THA. We are not the first study to examine outcomes following conversion HA to THA surgery. In 2002, Sierra and Cabanela [24] identified 132 HA to THA conversion surgeries. They reported a high rate of major perioperative complications (45%), including a 10% rate of dislocations, and a 9% rate of periprosthetic femoral fracture, all higher than the rates reported in this study. In a prospective analysis, Pankaj et al [26] reported on 44 cases of conversion of HA to THA. Similar to our study, aseptic loosening and prosthetic arthritis were the primary causes of HA failure and conversion HA to THA. Once again, they reported a higher rate of perioperative complications compared to our study (23% vs 11.7%), primarily due to a high rate of periprosthetic fractures (20%), which is a more technically difficult operation than revision surgery for aseptic femoral loosening. Newer surgical techniques or implant designs may also have led to a decrease in perioperative complication rates, which may account for the lower rates of perioperative complications seen in our study. Most recently, Taheriazam and Saeidinia [32] evaluated outcomes in 138 patients who underwent HA to THA. Once again, stem loosening and prosthetic arthritis were the primary causes for failed HA. Similar to our study, they report relatively low rates of major perioperative complications (9%), with no periprosthetic fractures. Given the lack of literature and to gain a better understanding of conversion HA to THA, we chose to compare this procedure to primary THA and revision THA. By comparing these 3 cohorts, we aimed at better understanding the perioperative characteristics of conversion HA to THA, allowing surgeons to more easily correlate the risks associated with this procedure. There are no clinical cohort studies to date that compared the outcomes of conversion HA to THA, primary THA, and revision THA. However, based on a similar premise, using the American College of Surgeons National Surgical Quality Improvement Project database, Schwarzkopf and Baghoolizadeh [33] compared outcomes between primary THA, conversion THA, and revision THA. This comparison was based on CPT codes, and, therefore, the conversion THA group was a combination of internal fixation to THA and HA to THA. Their study reported a large number of differences between conversion and
1172
N.O. Sarpong et al. / The Journal of Arthroplasty 34 (2019) 1168e1173
primary THA and found that conversion THA was most similar to revision THA. Our study was unique, in that we chose to isolate conversion HA to THA, a very different procedure compared to internal fixation conversion. Similar to their study, we report significant differences between a primary THA and a conversion HA to THA. Although a comparison between studies is difficult because of different outcome measures and statistical methods, overall, our study found more differences between conversion HA to THA and revision THA, particularly for orthopedic-specific complications at time points within and beyond 30 days postoperatively, which is not provided by the National Surgical Quality Improvement Project database. In other words, based on perioperative complications, conversion HA to THA falls more closely toward primary THA than revision THA, compared to the Schwarzkopf and Baghoolizadeh’s [33] conclusion that conversion THA better resembles revision THA. One potential hypothesis for this difference is that conversion of an HA to THA may be a less complex procedure than conversion of an ORIF (such as cephalomedullary nail). This study has a number of limitations. First, study size for our patient cohorts was relatively small, compared to larger registry studies. Unfortunately, conversion HA to THA does not have a specific CPT code, so identifying this patient population in registrybased studies is difficult. In addition, follow-up in this study was limited to the midterm, with a mean follow-up of 1.95, 2.80, 2.05 years for the primary, conversion HA to THA, and revision cohorts, respectively. This limitation restricted our ability to examine longterm survival rates for the 3 cohorts. Rates of revision surgery are important outcome measures, and our relatively short follow-up period restricted our ability to compare these rates among groups more long term. However, our follow-up period is greater than that in previous reports [33], and we were able to stratify by short term (90 days) and midterm (2 years). Finally, there are a number of additional cohort comparisons that may be relevant. For example, we chose to compare conversion HA to THA to primary THA and to revision THA of only the femoral component. Additional possible comparisons include conversion HA to THA with or without femoral component exchange to the other 2 cohorts, conversion HA to THA to revision both component THA, and other variations. We felt that the 3 cohort comparisons we chose was an appropriate balance of study feasibility and clinical relevance.
[3]
[4] [5]
[6]
[7] [8]
[9]
[10] [11]
[12]
[13]
[14] [15]
[16]
[17]
[18]
Conclusion
[19]
Conversion surgery to THA is a solution for failed HAs following femoral neck fractures. Our understanding of the outcomes and complications following this unique surgical procedure is limited. This is the first clinical study to compare complication and revision rates between conversion HA to THA, primary THA, and revision THA. In this study, conversion HA to THA more closely resembled primary THA than revision THA in overall complications within 2 years postoperatively. For operative time, blood loss, and periprosthetic dislocations, conversion HA to THA more closely resembled revision THA. By comparing conversion HA to THA to the more commonly performed procedures of primary THA and revision THA, the reported outcomes in this study can help surgeons appropriately stratify the risks involved in this surgery for this atrisk patient population.
[20]
[21]
[22]
[23]
[24] [25]
References [1] Bhandari M, Devereaux PJ, Tornetta P, Swiontkowski MF, Berry DJ, Haidukewych G, et al. Operative management of displaced femoral neck fractures in elderly patients. An international survey. J Bone Joint Surg Am 2005;87:2122e30. https://doi.org/10.2106/JBJS.E.00535. [2] Evaniew N, Madden K, Bhandari M. Cochrane in CORR®: arthroplasties (with and without bone cement) for proximal femoral fractures in adults. Clin
[26]
[27]
Orthop Relat Res 2014;472:1367e72. https://doi.org/10.1007/s11999-0133328-5. Ahn J, Man L-X, Park S, Sodl JF, Esterhai JL. Systematic review of cemented and uncemented hemiarthroplasty outcomes for femoral neck fractures. Clin Orthop Relat Res 2008;466:2513e8. https://doi.org/10.1007/s11999-0080368-3. Braithwaite RS, Col NF, Wong JB. Estimating hip fracture morbidity, mortality and costs. J Am Geriatr Soc 2003;51:364e70. Grosso MG, Danoff JR, Padgett DE, Iorio R, Macaulay WB. The cemented unipolar prosthesis for the management of displaced femoral neck fractures in the dependent osteopenic elderly. J Arthroplasty 2016;31:1040e6. https:// doi.org/10.1016/j.arth.2015.11.029. Lee BP, Berry DJ, Harmsen WS, Sim FH. Total hip arthroplasty for the treatment of an acute fracture of the femoral neck: long-term results. J Bone Joint Surg Am 1998;80:70e5. Bourne RB, Maloney WJ, Wright JG. An AOA critical issue. The outcome of the outcomes movement. J Bone Joint Surg Am 2004;86-A:633e40. Ong KL, Lau E, Suggs J, Kurtz SM, Manley MT. Risk of subsequent revision after primary and revision total joint arthroplasty. Clin Orthop Relat Res 2010;468: 3070e6. https://doi.org/10.1007/s11999-010-1399-0. Makarewich CA, Anderson MB, Gililland JM, Pelt CE, Peters CL. Ten-year survivorship of primary total hip arthroplasty in patients 30 years of age or younger. Bone Joint J 2018;100-B:867e74. https://doi.org/10.1302/0301620X.100B7.BJJ-2017-1603.R1. Pakos EE, Paschos NK, Xenakis TA. Long term outcomes of total hip arthroplasty in young patients under 30. Arch Bone Jt Surg 2014;2:157e62. Girard J, Bocquet D, Autissier G, Fouilleron N, Fron D, Migaud H. Metal-onmetal hip arthroplasty in patients thirty years of age or younger. J Bone Joint Surg Am 2010;92:2419e26. https://doi.org/10.2106/JBJS.I.01644. Chin G, Wright DJ, Snir N, Schwarzkopf R. Primary vs conversion total hip arthroplasty: a cost analysis. J Arthroplasty 2016;31:362e7. https://doi.org/ 10.1016/j.arth.2015.08.014. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 2007;89:780e5. https://doi.org/10.2106/JBJS.F.00222. Gullberg B, Johnell O, Kanis JA. World-wide projections for hip fracture. Osteoporos Int 1997;7:407e13. Burgers PT, Van Geene AR, Van den Bekerom MP, Van Lieshout EM, Blom B, Aleem IS, et al. Total hip arthroplasty versus hemiarthroplasty for displaced femoral neck fractures in the healthy elderly: a meta-analysis and systematic review of randomized trials. Int Orthop 2012;36:1549e60. https://doi.org/ 10.1007/s00264-012-1569-7. Keating JF, Grant A, Masson M, Scott NW, Forbes JF. Displaced intracapsular hip fractures in fit, older people: a randomised comparison of reduction and fixation, bipolar hemiarthroplasty and total hip arthroplasty. Health Technol Assess 2005;9:iiieiv. ixex, 1e65. van den Bekerom MP, Sierevelt IN, Bonke H, Raaymakers EL. The natural history of the hemiarthroplasty for displaced intracapsular femoral neck fractures. Acta Orthop 2013;84:555e60. https://doi.org/10.3109/ 17453674.2013.867763. Grosso MJ, Danoff JR, Murtaugh TS, Trofa DP, Sawires AN, Macaulay WB. Hemiarthroplasty for displaced femoral neck fractures in the elderly has a low conversion rate. J Arthroplasty 2017;32:150e4. https://doi.org/10.1016/ j.arth.2016.06.048. Wachtl SW, Jakob RP, Gautier E. Ten-year patient and prosthesis survival after unipolar hip hemiarthroplasty in female patients over 70 years old. J Arthroplasty 2003;18:587e91. Avery PP, Baker RP, Walton MJ, Rooker JC, Squires B, Gargan MF, et al. Total hip replacement and hemiarthroplasty in mobile, independent patients with a displaced intracapsular fracture of the femoral neck: a seven- to ten-year follow-up report of a prospective randomised controlled trial. J Bone Joint Surg Br 2011;93:1045e8. https://doi.org/10.1302/0301620X.93B8.27132. Alazzawi S, Sprenger De Rover WB, Brown J, Davis B. The conversion rate of bipolar hemiarthroplasty after a hip fracture to a total hip arthroplasty. Clin Orthop Surg 2012;4:117e20. https://doi.org/10.4055/cios.2012.4.2.117. Tanous T, Stephenson KW, Grecula MJ. Hip hemiarthroplasty after displaced femoral neck fracture: a survivorship analysis. Orthopedics 2010;33:385. https://doi.org/10.3928/01477447-20100429-08. Grosso MJ, Danoff JR, Thacher R, Murtaugh TS, Hickernell TR, Shah RP, et al. Risk factors for conversion surgery to total hip arthroplasty of a hemiarthroplasty performed for a femoral neck fracture. Hip Int 2018;28:168e72. https://doi.org/10.1177/1120700018768654. Sierra RJ, Cabanela ME. Conversion of failed hip hemiarthroplasties after femoral neck fractures. Clin Orthop Relat Res 2002;399:129e39. Figved W, Dybvik E, Frihagen F, Furnes O, Madsen JE, Havelin LI, et al. Conversion from failed hemiarthroplasty to total hip arthroplasty: a Norwegian Arthroplasty Register analysis of 595 hips with previous femoral neck fractures. Acta Orthop 2007;78:711e8. https://doi.org/10.1080/ 17453670710014473. Pankaj A, Malhotra R, Bhan S. Conversion of failed hemiarthroplasty to total hip arthroplasty: a short to mid-term follow-up study. Indian J Orthop 2008;42:294e300. https://doi.org/10.4103/0019-5413.41852. Fichman SG, M€ akinen TJ, Vincent A, Lozano B, Safir O, Kuzyk PRT. Complications following conversion of a hip hemiarthroplasty to a total hip
N.O. Sarpong et al. / The Journal of Arthroplasty 34 (2019) 1168e1173 arthroplasty. Int Orthop 2015;39:2335e9. https://doi.org/10.1007/s00264015-2827-2. [28] Sharkey PF, Rao R, Hozack WJ, Rothman RH, Carey C. Conversion of hemiarthroplasty to total hip arthroplasty: can groin pain be eliminated? J Arthroplasty 1998;13:627e30. [29] Sarmiento A, Gerard FM. Total hip arthroplasty for failed endoprostheses. Clin Orthop Relat Res 1978;137:112e7. [30] Sah AP, Estok DM. Dislocation rate after conversion from hip hemiarthroplasty to total hip arthroplasty. J Bone Joint Surg Am 2008;90:506e16. https:// doi.org/10.2106/JBJS.G.00479.
1173
[31] Park K-S, Chan C-K, Lee D-H, Yoon T-R. Midterm results of conversion from failed bipolar hemiarthroplasty to total hip arthroplasty. Indian J Orthop 2018;52:369e73. https://doi.org/10.4103/ortho.IJOrtho_ 494_16. [32] Taheriazam A, Saeidinia A. Conversion of failed hemiarthroplasty to total hip arthroplasty: a short-term follow-up study. Medicine (Baltimore) 2017;96: e8235. https://doi.org/10.1097/MD.0000000000008235. [33] Schwarzkopf R, Baghoolizadeh M. Conversion total hip arthroplasty: primary or revision total hip arthroplasty. World J Orthop 2015;6:750e3. https:// doi.org/10.5312/wjo.v6.i10.750.