Abstract
Aims
Two-stage exchange revision total hip arthroplasty (THA) performed in case of periprosthetic joint infection (PJI) has been considered for many years as being the gold standard for the treatment of chronic infection. However, over the past decade, there have been concerns about its safety and its effectiveness. The purposes of our study were to investigate our practice, collecting the overall spacer complications, and then to analyze their risk factors.
Methods
We retrospectively included 125 patients with chronic hip PJI who underwent a staged THA revision performed between January 2013 and December 2019. All spacer complications were systematically collected, and risk factors were analyzed. Statistical evaluations were performed using the Student's t-test, Mann-Whitney U test, and Fisher's exact test.
Results
Our staged exchange practice shows poor results, which means a 42% mechanical spacer failure rate, and a 20% recurrent infection rate over the two years average follow-up period. Moreover, we found a high rate of spacer dislocation (23%) and a low rate of spacer fracture (8%) compared to the previous literature. Our findings stress that the majority of spacer complications and failures is reflecting a population with high comorbid burden, highlighted by the American Society of Anesthesiology grade, Charlson Comorbidity Index, and Lee score associations, as well as the cardiac, pulmonary, kidney, or hepatic chronic conditions.
Conclusion
Our experience of a two-stage hip exchange revision noted important complication rates associated with high failure rates of polymethylmethacrylate spacers. These findings must be interpreted in the light of the patient’s comorbidity profiles, as the elective population for staged exchange has an increasing comorbid burden leading to poor results. In order to provide better results for this specific population, our conclusion suggests that comparative strategy studies are required to improve our therapeutic indication.
Cite this article: Bone Jt Open 2022;3(6):485–494.
Take home message
Our experience of a two-stage hip exchange revision reports important complication rates associated with high failure rates of polymethyl methacrylate spacers.
The elective population for staged exchange has an increasing comorbid burden which leads to poor results.
Introduction
Although total hip arthroplasty (THA) is a successful surgery in terms of alleviating pain and restoring functional activity in patients with advanced degenerative joint disease,1 periprosthetic joint infection (PJI) remains one of the most feared complications.
The consequences are disastrous, representing a worldwide economic burden estimated at $753.4 million on the American healthcare system alone.2 A devastating complication for patients, it may further severely limit joint function, and increases morbidity and mortality.3
Unfortunately, this complication is becoming more common, due to an increasing number of THAs4 and a persistent PJI rate.5 Indeed, despite continued progress and substantial efforts to develop preventive strategies, the rate of PJI continues to range between 1% and 2%.6
Nevertheless, despite the urge for effective strategies, the best treatment for chronic PJI remains controversial. Presently, two-stage exchange arthroplasty is the popular surgical treatment for the surgical management of PJI.7 However, to date, there are no randomized clinical trials that provide indications or contraindications for two-stage exchange arthroplasty.8,9 Additionally, there is a variability in the reported rates in specific complication,10-13 in morbidity and mortality,3,10,14 and success in eradicating infection.8,11,12,14
The recent literature has brought to light inferior clinical and perioperative outcomes as a result of mechanical complications, as well as a higher risk of reinfection,15 lower survivorship, and functional outcomes after spacer exchange.16 Moreover, given the substantial number of patients who never undergo reimplantation, the staged revision does not result to previously reported high rates of cure.14,17,18 Brown et al10 has reported mortality rates similar to prostatectomy or kidney transplant. Furthermore, the literature emphasizes the high rates of spacer retention,19 leading to frequent aseptic failure and poor outcome,20,21 along with the high rates of persistent infection,11,21 representing a dramatic scenario, leading to poor therapeutic possibilities.22,23
Based on these findings, this study was conducted in order to investigate our practice of a two-stage exchange strategy. Therefore, our purposes were to provide a complete picture of the overall spacer complications, and expand on this analysis by assessing the risk factors of two-stage exchange arthroplasty failure of our practice.
Methods
Patient demographic characteristics
Following institutional review board approval, we retrospectively retrieved records for 28,717 THA PJI from our database coding system who fit the criteria: removal of total hip prosthesis coupled with insertion of a cement spacer. Overall, 131 patients met the inclusion criteria for a two-stage exchange for THA PJI between the 1 January 2013 and 31 December 2019, performed in three university hospitals. After exclusions of the static spacers (four patients) and the failure of a previous two-stage procedure (two patients), 125 patients were included in the final cohort.
Primary outcome: overall of staged-exchange complication
A retrospective chart review was performed, identifying the general spacer complications, the specific spacer complications, and the failure of staged exchange procedure.
General complication
First, general complications have been reported, either as a medical or surgical complication. A medical complication was reported if a specific medical care was required while the patient’s postoperative hospital stay. The surgical complications were assessed using the five-level Clavien-Dindo classification,24 which assigns a score based on the importance of the treatment’s complication.
Specific spacer complication
Specific spacer complications were then analyzed, in the form of mechanical complications and spacer exchange for persistent infection. Mechanical complications were noted, based on radiological analysis or any reference in the medical record, and spacer exchange was reported based on the associated surgical report.
Failure of staged exchange procedure
Finally, two-stage revision failures were reported as the occurrence of death, spacer retention, and recurrent PJI. Death was reported during the interstage and after the second stage follow-up based on the medical record and a national database search. The retention of the spacer was defined as the inability to complete the second stage.
We stressed that recurrent PJI was judged according to the criteria published by Diaz-Ledezma et al25 after a Delphi based international and multidisciplinary consensus.
Secondary outcome: Independent significant risk factors for outcomes
From our statistical institutional database, queries were performed to retrieve patient medical records to collect host demographic characteristics and comorbidity. Moreover, we collected from the pre-anesthesic consultation, and clinical risk-stratification classification systems to provide an overall understanding of the patient’s health status, such as the American Society of Anesthesiologists (ASA) physical status classification to assess perioperative risk,26 and the Lee score for perioperative cardiac events.27
In addition, an age-adjusted Charlson Comorbidity Index (CCI) score28 was calculated using a standardized online medical calculator, to estimate mortality risk and disease burden over one-year.
Relying on a previous analysis of the literature, preoperative biological data from first stage and second stage, and radiological spacer data after first stage,29 were collected as potential complication risk factors,.
Finally, we reported the organism of PJI hip aspiration, and of the surgical samples from the first and second stages, based on the classification of organisms used by Rava et al9 in their systematic review of two-stage exchange procedures.
Statistical analysis
Descriptive statistics for spacer complications are presented as means and standard deviations (SDs) for continuous variables and as frequencies and percentages for categorical variables. Student's t-test and Mann-Whitney tests were used to compare groups. Univariate analysis using Fisher’s exact test was used to determine the association between spacer complication and independent risk factors. All statistical analysis were performed using the online software Easymedstat version 3.9,30 and significance was set at p < 0.05.
Results
Patient demographic and outcome characteristics
During the study period, 125 patients were treated with a two-stage revision THA; 62.4% (78/125) were male, mean age was 64.8 years (31 to 88;SD 12.2), with 37.9% (47/124) aged above 70 years, and mean BMI was 26.9 kg/m2 (17.70 to 44.9; SD 5.2). The mean CCI score was 4.1 (00 to 11; SD 2.5), and 26.1% of patients (33/124) had a score above 6. The mean ASA score was 2.4 (1 to 4; SD 0.8), 46.4% of patients (52/125) had a score above 3, and the mean Lee score was 0.5 (0 to 4; SD 0.8). Table I illustrates the patient’s demographic, clinical, and outcome characteristics.
Table I.
Patient demographic, clinical, and outcome characteristics.
| Variable | Data | Range | 95% CI |
|---|---|---|---|
| Demographic characteristics | |||
| Mean age, yrs | 65.93 | 31 to 88 | 63.757 to 68.095 |
| Age < 45 yrs, n (%) | 9 (7.44) | ||
| Age > 70 yrs, n (%) | 47 (37.9) | ||
| Mean BMI first stage | 26.92 | 17.70 to 44.90 | 25.844 to 27.992 |
| BMI > 30 kg/m2, n (%) | 26 (28.89) | ||
| Sex, n (%) | |||
| Male | 83 (62.4) | ||
| Female | 52 (36.8) | ||
| Medical history | |||
| Variable | Data | Variable | Result, n (%) |
| Mean CCI | 4.07 (2.529) | Alcohol use | |
| CCI ≥ 6, n (%) | 33 (26.61) | Active | 15 (12.61) |
| Sober | 4 (3.36) | ||
| None | 100 (84.03) | ||
| ASA score, mean (SD) | 2.43 (0.753) | Tobacco use | |
| 1 | 11 (9.82) | Active | 30 (25.0) |
| 2 | 49 (43.75) | Former smoker | 30 (25.0) |
| 3 | 45 (40.18) | None | 61 (50.0) |
| 4 | 7 (6.25) | ||
| 5 | 0 (0.0) | ||
| 6 | 0 (0.0) | ||
| Lee score, mean (SD) | 0.46 (0.760) | Chronic dermatological disease | 5 (4.17) |
| 0 | 69 (59.48) | ||
| 1 | 24 (20.69) | ||
| 2 | 8 (6.9) | ||
| 3 | 1 (0.86) | ||
| Allergic disposition, n (%) | 30 (28.04) | Inflammatory arthritis, n (%) | 6 (4.96) |
| Chronic pulmonary disease, n (%) | 19 (15.7) | Osteoporosis, n (%) | 4 (3.77) |
| Chronic cardiac disease, n (%) | 10 (8.26) | Dementia, n (%) | 9 (7.44) |
| Chronic liver disease, n (%) | 9 (7.5) | Hemiplegia, n (%) | 3 (2.52) |
| Chronic kidney disease, n (%) | 16 (13.22) | Psychiatric, n (%) | 22 (18.33) |
| HIV, n (%) | 3 (2.59) | Hemophilia, n (%) | 2 (1.67) |
| Diabetes, n (%) | 25 (21.05) | Malignancy, n (%) | 19 (15.57) |
| Implantable chamber, n (%) | 24 (20.34) | Immunosuppressive drugs, n (%) | 11 (9.17) |
| Pressure sore, n (%) | 6 (5.04) | Anticoagulant drugs | 16 (13.33) |
| Surgical history | |||
| Variable | Data | Range | 95% CI |
| Mean diagnostic delay, mnths | 8.29 | 0.00 to 96.00 | 5.606 to 10.968 |
| Mean time to reimplantation, mnths | 4.69 | 2.00 to 18.00 | 4.093 to 5.282 |
| Mean pevious hip surgeries | 2.81 | 1.00 to 6.00 | 2.556 to 3.059 |
| Previous DAIR, n (%) | 42 (60.87) | ||
| Macroscopic gross purulence, n (%) | 31 (36.05) | ||
| Extended trochanteric osteotomy, n (%) | 51 (55.43) | ||
| Sinus tract, n (%) | 36 (31.86) | ||
| Mean first stage operative time, mins | 152.86 | 90.00 to 240.00 | 120.456 to 185.258 |
| Mean first stage hospitalization time, days | 16.07 | 3.00 to 150.00 | 12.997 to 19.147 |
| First stage intensive care, n (%) | 6 (4.80) | ||
| Mean first stage blood transfusion | 1.34 (1.614) | ||
| Haematoma, n (%) | |||
| Total | 10 (8.06) | ||
| First stage | 5 (4.0) | ||
| Second stage | 3 (2.4) | ||
| Mean weightbearing | 0.29 | 0.00 to 1.00 | 0.192 to 0.384 |
| Mean second stage operative time, min | 143.08 | 60.00 to 210.00 | 115.857 to 170.297 |
| Mean second stage hospitalization time, days | 10.61 | 4.00 to 33.00 | 9.448 to 11.779 |
| Second stage intensive care | 4 (3.2) | ||
| Second stage blood transfusion | 1.17 | 0.00 to 8.00 | 0.788 to 1.545 |
| Radiological data | |||
| Mean leg length discrepancy, mm | 13.91 | -24.00 to 187.00 | 7.825 to 19.987 |
| Mean implant offset | 39.44 | 0.00 to 64.00 | 37.545 to 41.336 |
| Offset < 30 mm, n (%) | 10 (9.71) | ||
| Offset > 45 mm, n (%) | 20 (19.42) | ||
| Mean spacer’s length | 197.07 | 35.25 to 340.00 | 186.453 to 207.688 |
| Length < 150 mm, n (%) | 16 (14.29) | ||
| Mean mismatch head/acetabulum | 6.15 | 0.00 to 50.00 | 4.767 to 7.541 |
| Mismatch < 4 mm, n (%) | 50 (50.51) | ||
| Mismatch > 8 mm, n (%) | 25 (25.5) | ||
| Meam head/neck ratio | 2.04 | 1.09 to 3.29 | 1.968 to 2.113 |
| H/N < 1.7, n (%) | 19 (17.92) | ||
| H/N > 2.4, n (%) | 13 (12.15) | ||
| Biological data | |||
| First stage | |||
| Mean albumin, g/l | 34.16 | 19.20 to 46.10 | 31.712 to 36.616 |
| Mean Hba1c, % | 6.66 | 5.00 to 11.60 | 4.872 to 8.446 |
| Mean WBC | 8.53 | 1.90 to 24.00 | 7.792 to 9.277 |
| WBC > 12, g/dl, n (%) | 7 (8.24) | ||
| Mean PMN, G/l | 6.22 | 2.10 to 22.00 | 5.412 to 7.021 |
| Mean CRP | 58.64 | 3.00 to 343.50 | 42.420 to 74.856 |
| CRP > 70 mg/l, n (%) | 16 (20.25) | ||
| Mean ΔHb, g/dl | 2.65 | 0.00 to 11.90 | 2.234 to 3.060 |
| Second stage | |||
| Mean albumin, g/l | 38.12 | 24.20 to 48.70 | 34.919 to 41.319 |
| MeanWBC | 6.95 | 0.00 to 15.29 | 6.318 to 7.572 |
| WBC > 12 G/l, n (%) | 6 (7.59) | ||
| Mean PMN (G/l) | 5.03 | 0.00 to 12.32 | 4.401 to 5.655 |
| Mean CRP | 21.09 | 1.2 to 64.90 | 12.213 to 29.965 |
| CRP > 40 mg/l, n (%) | 7 (11.48) | ||
| Mean ΔHb, g/d | 2.78 | 0.00 to 10.70 | 2.431 to 3.132 |
-
CCI, Charlson Comorbidity Index; CI, confidence interval; CRP, C-reactive protein; DAIR, debridement, antibiotics, irrigation, and retention; Hb, haemoglobin; Hba1c, glycosylated haemoglobin; PMN, polymorphonuclear neutrophil; WBC, white blood cell.
The mean reimplantation time was 4.7 months (2 to 18; SD 3.1), and the mean follow-up was 2.1 years (0.4 to 6.04; SD 1.5). At the latest follow-up, we reported a 30.4% (38 patients) rate of lost to follow-up, which led to a search of a national database to reduce this result to 20% (25 patients). Figure 1 depicts the outcome of the final cohort.
Fig. 1
Flowchart depicting the outcomes of the final cohort.
The main infecting organisms in the diagnostic hip aspiration, were coagulase-negative staphylococcus (CoNS) in 34.02% cases (33/97), methicillin-sensitive Staphylococcus aureus (MSSA) in 29.9% cases (29/97), and gram-negative bacilli in 24.7% cases (24/97).
At the time of first stage, the most common organisms were CoNS in 34.07% of cases (31/91), MSSA in 31.87% cases (29/91), and gram-negative bacilli in 26.37% of cases (24/91). Figure 2 reports the overall organism at each stage of the procedure.
Fig. 2
Overall organism at each stage of the staged exchange procedure.
Primary outcome: Overall of staged-exchange complication
Table II summarizes the overall spacer complication for this study.
Table II.
Summary of the spacer complication in the cohort.
| Variable | General complication, n (%) | Spacer complication, n (%) | Two-stage exchange failure, n (%) | ||
|---|---|---|---|---|---|
| Spacer dislocation | 27 (23.08) | Recurrent PJI | 23 (21.3) | ||
| Clavien-Dindo global | Mean time, days | 34.39 (49.823) | |||
| 0 | 64 (51.61) | ||||
| 1 | 0 (0.0) | ||||
| 2 | 45 (36.25) | ||||
| 3 | 36.25 (0.81) | ||||
| 4 | 1 (8.87) | ||||
| 5 | 11 (2.42) | ||||
| Clavien-Dindo first stage | Femoral fracture | 31 (13.48) | Death | 4 (3.42) | |
| 0 | 4 (67.74) | First stage | 13 (26.27) | ||
| 1 | 0 (0.0) | Interstage | 6 (23.08) | ||
| 2 | 32 (25.81) | First stage + interstage | 20 (16.13) | ||
| 3 | 0 (0.0) | Second stage | 5 (19.23) | ||
| 4 | 6 (4.84) | ||||
| 5 | 2 (1.61) | ||||
| Clavien-Dindo second stage | Spacer fracture | 14 (11.29) | Spacer retention | 4 (3.42) | |
| 0 | 91 (73.39) | ||||
| 1 | 0 (0.0) | ||||
| 2 | 30 (24.19) | ||||
| 3 | 0 (0.0) | ||||
| 4 | 1 (0.87) | ||||
| 5 | 2 (51.61) | ||||
| Medical complication | 25 (20.16) | Acetabular fracture | 5 (4.35) | ||
| Antibiotic complication | 21 (14.0) | Spacer migration | 5 (4.10) | ||
| Antibiotic side-effects | 14 (11.29) | Spacer exchange | 20 (16.67) | ||
| Mean time, days | 220.94 (302.838) | ||||
| Antibiotic allergic reaction | 7 (5.6) | Time to reimplantation | |||
| 0 to 2 mnths | 14 (11.29) | ||||
| 2 to 4 mnths | 52 (49.06) | ||||
| 4 to 8 mnths | 29 (27.36) | ||||
| > 8 mnths | 11 (10.28) | ||||
General complication
First, for general complications, 20.2% (31/124) of medical complications, and a 49.2% (61/124) of surgical complications were reported. Table III presents the Clavien-Dindo complication grade at each stage of the procedure.
Table III.
Summary of the grades of complications according to Clavien-Dindo classification.
| Grade | Global, n (%) | First stage | Second stage |
|---|---|---|---|
| Minor (I-II) | 45 (36.0) | 32 (25.6) | 30 (24.0) |
| 0 | 64 (51.2) | 84 (67.2) | 91 (72.8) |
| I | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| II | 45 (36.0) | 32 (25.6) | 30 (24.0) |
| Major (III-IV-V) | 15 (12.0) | 8 (6.4) | 6 (4.8) |
| III | 1 (0.8) | 0 (0.0) | 0 (0.0) |
| IV | 11 (8.8) | 6 (4.8) | 1 (0.8) |
| V | 3 (2.4) | 2 (4.8) | 5 (4.0) |
Specific spacer complication
Second, concerning specific spacer complication, we report the occurrence of mechanical complications in 52 of 125 cases (41.6%), which were mostly spacer dislocations in 27/125 cases (23.08%), with a mean of 34.39 days, and femoral fractures in 31 of 125 cases (26.27%). Moreover, during the interstage, spacer exchange for persistent infection occurred in 20/120 cases (16.67%).
Failure of staged exchange procedure
Finally, we reported as failures of two-stage exchange, a 3.2% (4/125) death rate during the interstage period, and a 10.18% (11/108) death rate after reimplantation, for a mean follow-up of 15.6 months when taking into account the patients never reimplanted. In addition, we report an overall incidence of spacer retentions for the cohort of 3.4% (4/125), and of recurrent PJIs of 21.3% (23/118) when excluding the patients never reimplanted. Concerning the recurrent PJI, 21.7% (5/23) were considered as a persistent infection, presenting the same organism at the time of initial resection arthroplasty or previous hip aspiration, and 30.4% (7/23) were considered as a reinfection. Figure 3 provides a summary of the organism at each stage.
Fig. 3
Overall recurrent periprosthetic joint infecting organism.
Secondary outcome: Independent significant risk factors for outcomes
A full summary of the independent significant risk factors associated with our studied outcomes is reported in Table IV.
Table IV.
Summary of independent significant risk factors for studied outcomes.
| Risk factor | Odds ratio | p-value* |
|---|---|---|
| Medical complication | ||
| Age | 2.5 | 0.013 |
| ASA score | N/A | 0.027 |
| CCI | 2.99 | 0.017 |
| CCI > 6 | 2.64 | 0.034 |
| Chronic congestive heart failure | 5.44 | 0.012 |
| Chronic lung disease | 6.007 | 0.0007 |
| Dementia | 4.35 | 0.041 |
| Implantable chamber | 2.8 | 0.030 |
| WBC | 1.7 | 0.012 |
| WBC > 12 | 6.36 | 0.042 |
| PMN | 3.13 | 0.012 |
| Haemoglobin loss | 0.16 | 0.004 |
| Death | 9.86 | 0.047 |
| Clavien-Dindo | 16.97 | 0.0001 |
| Surgical complication | ||
| Age | 2.49 | 0.046 |
| Chronic lung disease | 7.03 | 0.001 |
| Medical complication | 16.97 | 0.0001 |
| Spacer dislocation | ||
| Offset < 30 mm | 5.83 | 0.012 |
| Spacer length < 150 mm | 4.86 | 0.011 |
| Femoral fracture | ||
| Osteoporosis | N/A | .0003 |
| Spacer exchange | ||
| ASA score | N/A | 0.012 |
| Chronic liver disease | 2.6 | 0.001 |
| First stage - sinus tract | 4.27 | 0.011 |
| Anticoagulant drug use | 4.06 | 0.023 |
| Implantable chamber | 4.27 | 0.006 |
| Time to reimplantation | N/A | 0.0005 |
| Hip aspiration - Gram-negative bacilli | 3.7 | 0.026 |
| First stage - Staphylococcus aureus | N/A | 0.030 |
| First stage - MRSA | 17.29 | 0.011 |
| First stage - antibiotic-resistant organism | 8.8 | 0.0008 |
| Medical complications | 16.97 | 0.024 |
| Recurrent PJI | 9.773 | 0.0001 |
| Time to reimplantation, 0 to 2 mnths | 0.1 | 0.008 |
| Time to reimplantation, > 8 mnths | 20.8 | 0.0001 |
| Recurrent PJI | ||
| ASA score | N/A | 0.009 |
| Chronic liver disease | 10.47 | 0.001 |
| Chronic dermatological disease | 7.18 | 0.047 |
| second stage - CRP> 40mg/l | 7.67 | 0.042 |
| Hip aspiration - gram-negative bacilli | 3.67 | 0.027 |
| First stage - drug-resistant organism | 7.22 | 0.002 |
| First stage - MRSA | 13.95 | 0.02 |
| First stage - Staphylococcus aureus | N/A | 0.03 |
| Spacer exchange | 9.77 | 0.0001 |
| Death | ||
| Age > 70 yrs | N/A | 0.015 |
| Lee score | N/A | 0.0009 |
| Chronic congestive heart failure | 12.35 | 0.036 |
| Anticoagulant drug use | 22.38 | 0.008 |
| First stage - haematoma | 39 | 0.007 |
| Medical complication | 11.74 | 0.033 |
| first stage - Clavien-Dindo | N/A | 0.0001 |
| Retained spacer | ||
| ASA score | N/A | 0.036 |
| Pressure sores | 27.5 | 0.012 |
| Dementia | 15.57 | 0.0279 |
-
*
Fisher's test was used to determine the association for each risk factor.
-
ASA, American Society of Anesthesiology; CCI, Charlson Comorbidity Index; CRP, C-reactive protein; MRSA, methicillin-resistant staphylococcus aureus; N/A, not applicable; PMN, polymorphonuclear neutrophil; WBC, white blood cell.
General complication
Concerning general complications, medical complications were associated with age (odds ratio (OR) 2.994; p < 0.010), ASA score (p < 0.027), CCI (OR 2.994; p < 0.010), CCI > 6 (OR 2.636; p < 0.034), chronic congestive heart failure (OR 5.438; p < 0.01), chronic lung disease (OR 6.007; p < 0.0007), and death (OR 9.86; p < 0.047, all Fisher's exact test). Surgical complications were associated with age (OR 2.492; p < 0.046, Fisher's exact test)) and chronic lung disease (OR 7.03; p < 0.001, Fisher's exact test).
Specific spacer complication
Regarding the specific spacer complication, our study found a significant association between the occurrence of a spacer dislocation and an offset < 30 mm (OR, 5.83; p < 0.012, Fisher's exact test), as well for a spacer length < 150 mm (OR 4.86; p < 0.011, Fisher's exact test).
Furthermore, the main risk factors significantly associated with an increased risk of spacer exchange were on the one hand: ASA score (p < 0.012), chronic liver disease (OR 2.6; p < 0.001), time to reimplantation > 8 months (OR 20.88; p < 0.0001, all Fisher's exact test).
On another hand, we recorded as increased risk factors the presence in the hip aspiration of a gram-negative bacilli (OR 3.7; p < 0.026), or in the first stage’s surgical samples of methicillin-resistant Staphylococcus aureus (MRSA) (OR 17.294; p < 0.012), or a drug-resistant organism (OR 8.8; p < 0.0008, all Fisher's exact test).
Failure of staged exchange procedure
Finally, concerning two-stage exchange failure, we report for the occurrence of death during the interstage, significant associations with age > 70 years (p < 0.015), Lee score (p < 0.0009), chronic congestive heart failure (OR 12.354; p < 0.036), anticoagulant drug use (OR 22.385; p < 0.008) and haematoma after first stage (OR 39.0; p < 0.007, all Fisher's exact test).
As well, we report significant independent risk factors for spacer retention, including; ASA score (p < 0.036), pressure sores (OR 27.5; p < 0.012), and dementia (OR 15.571; p < 0.0279, all Fisher's exact test).
Additionally, we point out in one hand, several independent significant risks for recurrent PJI as, ASA score (OR 2.338; p < 0.009, Fisher's exact test) and chronic liver disease (OR 10.471; p < 0.001, Fisher's exact test).
And on the other hand, the presence in the diagnostic PJI hip aspiration of a gram-negative bacilli (OR 3.674; p < 0.027) and the presence in the first stage’s surgical samples of a MRSA (OR 13.95; p < 0.02) or a drug resistant organism (OR 7.22; p < 0.002, all Fisher's exact test).
Discussion
The review of our practice of a two-stage revision THA for PJI emphasizes a procedure with a high risk of general and specific spacer complications, as well as high-staged exchange procedure failures.
Primary outcome: overall of staged-exchange complication
General complication
Concerning the general complication, we demonstrated a high rate of medical complication located in the high range reported in the literature, with results ranging from 8.8% to 46.3%.14,31,32 In addition, our study found a significant rate of surgical complications of 48.8%, compared with an average rate of 35.2% for global orthopaedic surgeries.24
Specific spacer complication
Regarding the specific spacer complication, we observed an overall spacer-related mechanical complication rate of 41.6%, also within the higher literature range, for results ranging from 19.6% to 40.8%.31,33 Our spacer dislocation rate of 23.1% appears to be consistent with the literature,9 although we reported more femoral fractures and a lower rate of spacer fractures.29,33 Moreover, we reported a higher spacer exchange rate of 16.67% than the literature’s results, ranging from 5% to 14%;19 nevertheless, the reason behind this difference remains unclear.
Failure of staged exchange procedure
For two-stage revision failures, we reported a mortality rate of 3.2%, which is consistent with the literature’s rates ranging from 2.6% to 7%.14,19 Furthermore, after a follow-up of 15.6 months, our death rate of 8.8% was similar to the literature’s rates,3,14,19 ranging from 6.5% to 10.6% after a one year follow-up.
Additionally, our study reported a 3.2% retention rate, while the literature evidenced a wide range of rate ranging from 13.5% to 68%,19-21 which underlies the lack of consensual definition.
Finally, we reported a relatively high recurrent PJI rate of 21.3% compared to the 10.4% rate reported by Lange et al8 in their meta-analysis. Despite the use of a consensual definition,25 we assume that this high rate results from the exclusion of the patients never reimplanted, and the inclusion of the patients considered as reinfected.
Secondary outcome: independent significant risk factors for outcomes
Regarding the risk factor analysis, our study reports a specific patient comorbidity profile, based on age, clinical risk-stratification systems, and chronic organ failure, which appears to be significantly associated with substantial staged exchange complication and failures.
The existing literature supports our conclusion, as spacer exchange is reported to be associated with CCI score,16 and chronic kidney and liver diseases.19 Moreover, concerning the occurrence of death within one year, Cancienne et al19 were able to identify an age above 85 years, liver, cardiac, and pulmonary diseases as risk factors. Additionally, spacer retention appears to be associated with advanced age, ASA grade, CCI score,21 and congestive heart failure.19 Finally, as well for recurrent PJI published data, association was found with the CCI,34 McPherson C3 score,12 and heart diseases.11
Taking into account these results, it appears that high comorbid burden among patients undergoing a staged exchange procedure leads to poor results.
Meanwhile, the elective arthroplasty patient population is ageing and is associated with increasing comorbidity indices,4 leading to a similar change in the staged revision population profile,35 as evidenced by our study, with a 40% rate of patients aged above 70 years, and half our population with an ASA grade above 3.
Thus, the fair results reported in this analysis regarding our practice of staged exchange arthroplasty for PJI need to take into account the high comorbidity profile of our population. As a result, our study highlights the need to consider specific care concerning chronic PJI for this specific population in order to provide better results.
Limitations
There are several limitations in this study, many of which common to most of the studies. First, the retrospective nature leads to unavoidable memory bias and data loss. Second, we acknowledge a high rate of lost to follow-up that might underestimate our results, despite a national database search to limit bias. Finally, our multicentric study design introduces a heterogeneity of practice. However, we do not believe that this weakens our findings, as all practices have been standardized by a reference centre for the management of complex bone and joint infections.
Despite the aforementioned limitations, the present study is the first to the best of our knowledge to provide a complete picture of the overall clinical impact of PJI in total joint replacment in order to evaluate the complex interplay between risk factors and the outcome of this procedure.19
In conclusion, the analysis of our practice reports a high-risk procedure, with a 20% rate of medical complications, a 49% rate of surgical complications, a 42% rate of mechanical complications, and a high risk of failure for a quarter of our study cohort.
Our findings identify the age, the clinical risk-stratification systems, and chronic organ failure as key predictors of high risk of complication and failure during a two-stage procedure. Therefore, these poor results should be interpreted with caution, reflecting an increasing comorbidity burden among our population with chronic PJI.
As pointed out in our study, the increase in chronic PJI THA in a population of elective arthroplasty patients with higher medical comorbidities seems to be a new challenge for orthopaedic surgeons. Our results imply that further studies comparing different strategies for chronic PJI THA in this specific population are required to improve our therapeutic indications and provide better results.
References
1. Learmonth ID , Young C , Rorabeck C . The operation of the century: total hip replacement . Lancet . 2007 ; 370 ( 9597 ): 1508 – 1519 . Crossref PubMed Google Scholar
2. Premkumar A , Kolin DA , Farley KX , et al. Projected economic burden of periprosthetic joint infection of the hip and knee in the United States . J Arthroplasty . 2021 ; 36 ( 5 ): 1484 – 1489 . Crossref PubMed Google Scholar
3. Zmistowski B , Karam JA , Durinka JB , Casper DS , Parvizi J . Periprosthetic joint infection increases the risk of one-year mortality . J Bone Joint Surg Am . 2013 ; 95-A ( 24 ): 2177 – 2184 . Crossref PubMed Google Scholar
4. Cnudde P , Nemes S , Bülow E , et al. Trends in hip replacements between 1999 and 2012 in Sweden . J Orthop Res . 2018 ; 36 ( 1 ): 432 – 442 . Crossref PubMed Google Scholar
5. Gwam CU , Mistry JB , Mohamed NS , et al. Current epidemiology of revision total hip arthroplasty in the United States: national inpatient sample 2009 to 2013 . J Arthroplasty . 2017 ; 32 ( 7 ): 2088 – 2092 . Crossref PubMed Google Scholar
6. Jämsen E , Furnes O , Engesaeter LB , et al. Prevention of deep infection in joint replacement surgery . Acta Orthop . 2010 ; 81 ( 6 ): 660 – 666 . Crossref PubMed Google Scholar
7. Osmon DR , Berbari EF , Berendt AR , et al. Diagnosis and management of prosthetic joint infection: clinical practice guidelines by the Infectious Diseases Society of America . Clin Infect Dis . 2013 ; 56 ( 1 ): e1 – e25 . Crossref PubMed Google Scholar
8. Lange J , Troelsen A , Thomsen RW , Søballe K . Chronic infections in hip arthroplasties: comparing risk of reinfection following one-stage and two-stage revision: a systematic review and meta-analysis . Clin Epidemiol . 2012 ; 4 : 57 – 73 . Crossref PubMed Google Scholar
9. Rava A , Bruzzone M , Cottino U , Enrietti E , Rossi R . Hip spacers in two-stage revision for periprosthetic joint infection: a review of literature . Joints . 2019 ; 7 ( 2 ): 56 – 63 . Crossref PubMed Google Scholar
10. Browne JA , Cancienne JM , Novicoff WM , Werner BC . Removal of an infected hip arthroplasty is a high-risk surgery: putting morbidity into context with other major nonorthopedic operations . J Arthroplasty . 2017 ; 32 ( 9 ): 2834 – 2841 . Crossref PubMed Google Scholar
11. Triantafyllopoulos GK , Memtsoudis SG , Zhang W , Ma Y , Sculco TP , Poultsides LA . Periprosthetic infection recurrence after 2-stage exchange arthroplasty: failure or fate? J Arthroplasty . 2017 ; 32 ( 2 ): 526 – 531 . Crossref Google Scholar
12. Chalmers BP , Mabry TM , Abdel MP , Berry DJ , Hanssen AD , Perry KI . Two-stage revision total hip arthroplasty with a specific articulating antibiotic spacer design: reliable periprosthetic joint infection eradication and functional improvement . J Arthroplasty . 2018 ; 33 ( 12 ): 3746 – 3753 . Crossref PubMed Google Scholar
13. Garceau S , Warschawski Y , Sanders E , Gross A , Safir O , Kuzyk P . Impact of hip antibiotic spacer dislocation on final implant position and outcomes . J Arthroplasty . 2019 ; 34 ( 9 ): 2107 – 2110 . Crossref PubMed Google Scholar
14. Berend KR , Lombardi AV , Morris MJ , Bergeson AG , Adams JB , Sneller MA . Two-stage treatment of hip periprosthetic joint infection is associated with a high rate of infection control but high mortality . Clin Orthop Relat Res . 2013 ; 471 ( 2 ): 510 – 518 . Crossref PubMed Google Scholar
15. Tan TL , Goswami K , Kheir MM , Xu C , Wang Q , Parvizi J . Surgical treatment of chronic periprosthetic joint infection: fate of spacer exchanges . J Arthroplasty . 2019 ; 34 ( 9 ): 2085 – 2090 . Crossref PubMed Google Scholar
16. Klemt C , Smith EJ , Tirumala V , Bounajem G , van den Kieboom J , Kwon Y-M . Outcomes and risk factors associated with 2-stage reimplantation requiring an interim spacer exchange for periprosthetic joint infection . J Arthroplasty . 2021 ; 36 ( 3 ): 1094 – 1100 . Crossref PubMed Google Scholar
17. Cancienne JM , Granadillo VA , Patel KJ , Werner BC , Browne JA . Risk factors for repeat debridement, spacer retention, amputation, arthrodesis, and mortality after removal of an infected total knee arthroplasty with spacer placement . J Arthroplasty . 2018 ; 33 ( 2 ): 515 – 520 . Crossref PubMed Google Scholar
18. Choi HR , Freiberg AA , Malchau H , Rubash HE , Kwon YM . The fate of unplanned retention of prosthetic articulating spacers for infected total hip and total knee arthroplasty . J Arthroplasty . 2014 ; 29 ( 4 ): 690 – 693 . Crossref PubMed Google Scholar
19. Cancienne JM , Werner BC , Bolarinwa SA , Browne JA . Removal of an infected total hip arthroplasty: risk factors for repeat debridement, long-term spacer retention, and mortality . J Arthroplasty . 2017 ; 32 ( 8 ): 2519 – 2522 . Crossref PubMed Google Scholar
20. Wang Q , Goswami K , Kuo FC , Xu C , Tan TL , Parvizi J . Two-stage exchange arthroplasty for periprosthetic joint infection: the rate and reason for the attrition after the first stage . J Arthroplasty . 2019 ; 34 ( 11 ): 2749 – 2756 . Crossref PubMed Google Scholar
21. Petis SM , Perry KI , Pagnano MW , Berry DJ , Hanssen AD , Abdel MP . Retained antibiotic spacers after total hip and knee arthroplasty resections: high complication rates . J Arthroplasty . 2017 ; 32 ( 11 ): 3510 – 3518 . Crossref PubMed Google Scholar
22. Brown TS , Fehring KA , Ollivier M , Mabry TM , Hanssen AD , Abdel MP . Repeat two-stage exchange arthroplasty for prosthetic hip re-infection . Bone Joint J . 2018 ; 100-B ( 9 ): 1157 – 1161 . Crossref PubMed Google Scholar
23. Fagotti L , Tatka J , Salles MJC , Queiroz MC . Risk factors and treatment options for failure of a two-stage exchange . Curr Rev Musculoskelet Med . 2018 ; 11 ( 3 ): 420 – 427 . Crossref PubMed Google Scholar
24. Camino Willhuber G , Slullitel P , Taype Zamboni D , et al. Validation of a modified Clavien-Dindo Classification for postoperative complications in orthopedic surgery . Rev Fac Cien Med Univ Nac Cordoba . 2020 ; 77 ( 3 ): 161 – 167 . Crossref PubMed Google Scholar
25. Diaz-Ledezma C , Higuera CA , Parvizi J . Success after treatment of periprosthetic joint infection: a Delphi-based international multidisciplinary consensus . Clin Orthop Relat Res . 2013 ; 471 ( 7 ): 2374 – 2382 . Crossref PubMed Google Scholar
26. Doyle DJ , Goyal A , Garmon EH . American Society of Anesthesiologists Classification . StatPearls Publishing , 2022 . Google Scholar
27. Lee TH , Marcantonio ER , Mangione CM , et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery . Circulation . 1999 ; 100 ( 10 ): 1043 – 1049 . Crossref PubMed Google Scholar
28. Charlson ME , Pompei P , Ales KL , MacKenzie CR . A new method of classifying prognostic comorbidity in longitudinal studies: development and validation . J Chronic Dis . 1987 ; 40 ( 5 ): 373 – 383 . Crossref PubMed Google Scholar
29. Jones CW , Selemon N , Nocon A , Bostrom M , Westrich G , Sculco PK . The influence of spacer design on the rate of complications in two-stage revision hip arthroplasty . J Arthroplasty . 2019 ; 34 ( 6 ): 1201 – 1206 . Crossref PubMed Google Scholar
30. No authors listed . EasyMedStat ; . https://www.easymedstat.com ( date last accessed 20 May 2022 ). Google Scholar
31. Jung J , Schmid NV , Kelm J , Schmitt E , Anagnostakos K . Complications after spacer implantation in the treatment of hip joint infections . Int J Med Sci . 2009 ; 6 ( 5 ): 265 – 273 . Crossref PubMed Google Scholar
32. Cabrita HB , Croci AT , Camargo OP de , Lima ALLM de . Prospective study of the treatment of infected hip arthroplasties with or without the use of an antibiotic-loaded cement spacer . Clinics (Sao Paulo) . 2007 ; 62 ( 2 ): 99 – 108 . Crossref Google Scholar
33. Faschingbauer M , Reichel H , Bieger R , Kappe T . Mechanical complications with one hundred and thirty eight (antibiotic-laden) cement spacers in the treatment of periprosthetic infection after total hip arthroplasty . Int Orthop . 2015 ; 39 ( 5 ): 989 – 994 . Crossref PubMed Google Scholar
34. Zmistowski B , Tetreault MW , Alijanipour P , Chen AF , Della Valle CJ , Parvizi J . Recurrent periprosthetic joint infection: persistent or new infection? J Arthroplasty . 2013 ; 28 ( 9 ): 1486 – 1489 . Crossref Google Scholar
35. Chang CH , Lee SH , Lin YC , Wang YC , Chang CJ , Hsieh PH . Increased periprosthetic hip and knee infection projected from 2014 to 2035 in Taiwan . J Infect Public Health . 2020 ; 13 ( 11 ): 1768 – 1773 . Crossref PubMed Google Scholar
Author contributions
M. Jaubert: Writing – original draft.
M. Le Baron: Writing – original draft.
C. Jacquet: Investigation.
A. Couvreur: Investigation.
M. Fabre-Aubrespy: Resources.
X. Flecher: Writing – review & editing.
M. Ollivier: Supervision.
J-N. Argenson: Writing – review & editing.
Funding statement
The author(s) received no financial or material support for the research, authorship, and/or publication of this article.
ICMJE COI statement
The authors declare the following, all of which is unrelated to this work: J-N. Argenson reports royalties or licenses from Zimmer Biomet, and being the past-president of EKS, deputy editor of JBJS Am, and being chair of the APCORT board of trustees; M. Fabre-Aubrespy declares payments from Convatec for educational events; and X. Flecher reports royalties from EUROS, consulting fees from Styrker, and payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing, or educational events from Zimmer Biomet, and participation on a data safety monitoring board or advisory board for Corin and DePuy Synthes.
Open access funding
The authors report that the open access funding for this manuscript was self-funded.
© 2022 Author(s) et al.This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/
