With the demand for total joint arthroplasty and overall life expectancy increasing, there will be an increase in the need for revision arthroplasty surgeries. Given that revision joint surgeries are more demanding for both surgeon and patient with longer operative times, increased blood loss, and multiple patient comorbidities, the current mindset is that older patients who undergo a total hip revision or total knee revision have higher mortality rates than younger patients. We identified 1737 revision total joint patients who were at least 2 years postoperative for inclusion in the study. The overall perioperative mortality rate (defined as deaths occurring between 0 and 3 months following revision joint surgery) was calculated and then stratified by revision knee surgery, revision hip surgery, and age. In addition, mortality rates were compared for patients younger than 70 years, between 70 and 80 years and older than 80 years. The overall perioperative mortality rate after revision total hip or knee surgery was 0.7%. After stratifying by age, the perioperative mortality rate was 0.2% in patients younger than 70 years, 0.8% in patients 70 to 79 years, and 2.63% in patients older than 80 years. Of the 1737 patients, 541 died >1 year following their revision surgery at an average time to death of 6.9 years. The observed perioperative mortality rates following revision total joint surgery at a single center were extremely low among all age groups. Therefore, the age of patients undergoing revision surgery should not be the sole determinant of perioperative survival. Additionally, it appears that the mean postoperative survival noted here seems to justify the additional resources used in revision surgery regardless of age. As limited resources exert pressure on an already overburdened healthcare system, rationing of care for certain procedures may ensue using age as a specific criteria. This study should add clarity to this issue.
The demand for total joint arthroplasty is increasing exponentially. Coincident with this increase is an exponential increase in the number of revision surgeries performed. It is projected that the annual number of revision total hip arthroplasties (THAs) will increase 137% by the year 2030 and 601% for revision total knee arthroplasty (TKA).1,2 It is also expected that the number of adults older than age 85 will increase fourfold by the year 2050.1
With the demand for total joint arthroplasty and life expectancy increasing, there will inevitably be an increase in the need for revision arthroplasty surgeries in the future. Compared to primary surgery, these procedures are technically more demanding with longer operative times and increased blood loss.3 These factors, plus the prevalence of comorbidities, may increase the complication risk for vulnerable elderly patients. Since additional resources are used for revision procedures such as increased implant costs, increased operative time, and increased surgeon effort, the risk-benefit analysis of performing this type of surgery on elderly patients must be carefully considered.
It is frequently surmised that elderly patients undergoing revision total joint replacement have significantly higher mortality rates than their younger counterparts. However, there is insufficient empirical data to determine whether an absolute age can be established where the perioperative risks outweigh the clinical benefits of revision surgery.4 We sought to answer the following research questions: Is there an absolute age beyond which the perioperative risks outweigh the clinical benefits of revision surgery? Does the postoperative survival justify the additional resources used in revision surgery?
Materials and Methods
A query of our institutional prospective database was performed to document procedure type, patient date of birth, and date of surgery. The Social Security Death Index database was used to verify death and determine date of death. A total of 1737 revision total joint patients who had surgery between May 1986 and August 2005 were identified and included in this study.
The overall perioperative mortality rate was determined. Perioperative mortality was stratified by revision hip arthroplasty, revision knee arthroplasty, and age. The perioperative mortality rate was calculated by dividing the number of deaths occurring between 0 and 3 months following revision joint surgery by the total number of revision patients and reported as a percentage. The time to death was calculated and divided into 3 periods: 0 to 3 months (perioperative), 3 to 12 months (early postoperative), and >12 months (postoperative). The mortality rate within each time interval was determined and stratified by revision TKA and revision THA patients. Additionally, mortality rates were compared in patients younger than 70 years and older than 70 years at the time of surgery. To determine if there was an age threshold of prohibitive perioperative risk and unwarranted resource utilization, perioperative mortality rates, stratified by procedure type, were compared for patients younger than 70 years, older than 70 years and older than 80 years. Descriptive statistics were reported as frequencies and percentages.
The overall perioperative mortality rate following revision total joint surgery was 0.7% (12 of 1737). The perioperative mortality rate for revision TKAs was 0.3% (2 of 640) and 0.9% (10 of 1097) for revision THAs. Of the 1737 revision total joint patients, 640 underwent revision total knee surgery and 1097 patients underwent revision total hip surgery. Two revision TKA and 10 revision THA perioperative deaths occurred during the 3 months following surgery.
The overall early postoperative (3-12 months postoperative) mortality rate for revision joint replacements was 0.86% (23 of 1737). For revision TKAs the early postoperative mortality rate was 1.25% (8 of 64) compared to 1.36% (15 of 1097) for revision THAs.
The overall perioperative mortality was 0.20% (2 of 1024) in patients younger than 70 years, 0.8% (4 of 485) in patients 70 to 79 years, and 2.63% (6 of 228) in patients >80 years.
The perioperative mortality rate among revision hip patients younger than 70 years was 0.15% (1 of 652) compared to 1.03% (3 of 290) for revision hip patients between 70 and 79 years. For revision THA patients 80 years and older, the perioperative mortality rate increased to 3.87% (6 of 155).
While there was a slight increase in perioperative mortality rates for revision THA patients when stratified by age, this trend was not realized for revision TKA patients. The perioperative mortality rate among revision knee patients younger than 70 years was 0.27% (1 of 372) and 0.51% (1 of 195) for revision TKA patients between 70 and 79 years. Of the 73 revision TKA patients 80 years and older, there were no deaths within 0 to 3 months postoperatively.
Five hundred and forty one of 1737 (31%) revision joint patients died at >1 year following revision surgery at an average time to death of 6.9 years (1.2 to 19.4 years). Of the 640 revision TKA patients, 189 (29.5%) died following revision knee surgery at an average time to death of 6.8 years (1.8 to 19.4). Of the 1,097 THA patients, 352 (32.1%) died following revision hip surgery at an average time to death of 7.0 years (1.2 to 18.7 years).
Mortality risk following a surgical intervention is an important consideration for surgeon and patient alike. Revision total joint surgery is most commonly performed on elderly patients with multiple comorbidities. These procedures are often lengthy and involve significant blood loss. Therefore information concerning the mortality risk performing this procedure in an elderly population is particularly germane.
Using Medicare population data, Mahomed et al5 reported the 90-day mortality rate following revision surgery.5,6 In evaluating 13483 revision hip replacements and 11726 revision knee replacements, the 90-day mortality rates were 2.6% and 1.1% respectively. The authors also compared their mortality rates to the expected number of deaths in a comparable Medicare sample population. The overall standard mortality rate in the revision THR group was 1.9 (95% confidence interval (CI), 1.7-2.2) indicating that the 90-day mortality rate was almost double that of a comparable Medicare cohort that had not undergone revision THR.6 After revision TKR, the 90-day standardized mortality ratio was only 0.9 (95% CI, 0.7-1.0) indicating essentially equivalent or slightly better survival than was found in a Medicare cohort that had not undergone revision TKA.5
We were encouraged that our overall perioperative mortality compared favorably to that reported in the above series. Our overall perioperative mortality for revision surgery was only 0.7%. Our perioperative mortality for revision knee surgery was 0.3% compared to 1.1% in Mahomed et al.5 Our perioperative mortality for revision hip surgery was 0.9% compared to 2.6% according to Mahomed et al.6 The nearly 3 to 3.5×difference in mortality rates between a tertiary referral center such as ours and Medicare claims data from a general orthopedic population is difficult to explain. While these differences appear significant, patient mix and selection bias may have influenced these results making comparisons difficult. However, since tertiary referral centers tend to attract the most difficult revision cases in patients with the greatest number of comorbidites, one would surmise that their complication rates and mortality rates should in fact be higher than the Medicare claims data. This discrepancy in mortality outcomes may be explained by hospital and surgeon procedure volume. Katz et al7 evaluated the association between hospital and surgeon procedure volume and outcome in 58521 primary and 12956 revision hip arthroplasties. They concluded that patients treated at hospitals and by surgeons with higher annual caseloads of primary and revision THR had lower rates of mortality.7 Hervey at al8 evaluated 23,180 revision TKAs using the Healthcare Cost and Utilization Project Nationwide Inpatient Sample (HCUP) and demonstrated that higher surgeon and hospital volumes were significantly associated with lower risks of mortality.
Perioperative mortality following revision TKA was negligible regardless of age (2/640, 0.3%) with no deaths in the 73 revision knees performed in patients older than 80 years. However, in the revision hip group there appears to be an increased risk of perioperative mortality with increasing age. Only 11 revision hip patients (0.15%) younger than 70 years died in the perioperative period, while 2.02% of those patients older than 70 years, and 3.87% of patients older than 80 years died in the perioperative period.
While we did not have a nonoperative Medicare cohort with which to compare, our findings are consistent with those of Mahomed et al5 who found a higher standard mortality ratio with older age in the revision hip cohort. In their study, the age specific standard mortality ratios were 1.4 for the 65 to 69 year age group and 2.0 for the 80 to 84 year age group compared to a Medicare cohort that had not undergone revision hip arthroplasty. This relationship might be explained by the observation that physiologic reserve decreased with age and hence the risk of death increased with age in this revision hip group.6
One purpose of the present study was to address whether an absolute age can be established where the perioperative risk outweighs the clinical benefits of revision surgery. We have shown that the risk of death in the perioperative period following revision knee surgery is nominal, while the risk of death in the perioperative period following revision hip surgery appears acceptable regardless of age. While any death in the perioperative period is disturbing, the fact remains that even in those patients having revision hip surgery at an age older than 80 years, the chance of surviving was >96%.
Our second research question hoped to answer whether postoperative survival justified the additional resources used in revision surgery. Cost utility analysis is a type of economic analysis that is used to assess the value of an intervention in terms of improving both quality and quantity of life.9-11 By combining quality and duration of life into a single metric, the quality adjusted life year (QALY) allows for comparison of interventions.9 A cost utility ratio of less than $20,000 per QALY is considered an excellent value that is cost effective. A cost utility ratio between $20,000 and $100,000 per QALY is considered to be a good value, while a cost utility ratio of >$100,000 per QALY is generally considered not the best value for the resources spent.
Brauer et al9 found 6 cost utility ratios for comparison between THA and the alternative of observation. Each of these indicated that total joint arthroplasty was cost-effective below a threshold of $50 000 per quality adjusted life year. Since the mean hospital cost of revision surgery calculated from 1 study was $31 341, it appears that revision surgery is far below the threshold of a cost-effective intervention.3
Additionally the mean postoperative survival after revision surgery was approximately 7 years in both revision hip and knee cohorts showing that after spending approximately $31,000 on each patient for surgery, we had a cost of $4500 per QALY. While we have no way of quantifying the quality of life these deceased patients lived during those subsequent 7 years, the resource expenditures appear far below the threshold for cost effectiveness in quality adjusted life years.
Furthermore, the cost of revision surgery to the health care system is significantly lower than the expenditures of nursing home care for these patients if nonoperative care prevents ambulation. A 1991 study compared the cost of THA to maintaining a nonindependent person in a nursing home. They estimated this cost to be $30,000 per year.12 In inflation adjusted dollars, this cost would be $46,000 today. Therefore 1 year of nursing home care is significantly higher than the hospital cost associated with revision surgery ($46,000 vs $31,000).
The limitations of this study are those inherent in its retrospective design. While we determined this retrospective cohort using an established comprehensive prospective registry and verified this data with the Social Security Death Index database, it is possible that this method may have undercalculated the number of patients who were deceased.
As part of any preoperative screening process, the evaluating surgeon makes a risk benefit decision taking age, comorbidities, and technical issues into consideration. Admittedly, selection bias occurred in the preoperative evaluation process performed by a variety of revision surgeons at our tertiary referral center. The reader must understand this unavoidable limitation when interpreting this data.
We have shown that in carefully selected patients, perioperative mortality following revision surgery performed at a tertiary center is extremely low regardless of age. Surgeons should recognize that comorbid conditions and not age probably play the dominant role in patient survival. Advanced age should not be used as a sole determinant of perioperative survival. Additionally, it appears that the mean postoperative survival noted here seems to justify the additional resources used in revision surgery regardless of age. As limited resources exert pressure on an already overburdened healthcare system, rationing of care for certain procedures may ensue using age as a specific criteria. This study should add clarity to this issue.
- Kurtz SM, Ong KL, Schmier J, et al. Future clinical and economic impact of revision total hip and knee arthroplasty. J Bone Joint Surg Am. 2007; 89 Suppl 3:144-151.
- Kurtz S, Ong K, Lau E, et al. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am. 2007; 89(4):780-785.
- Bozic KJ, Katz P, Cisternas M, Ono L, Ries MD, Showstack J. Hospital resource utilization for primary and revision total hip arthroplasty. J Bone Joint Surg Am. 2005; 87(3):570-576.
- Parvizi J, Pour AE, Keshavarzi NR, DApuzzo M, Sharkey PF, Hozack WJ. Revision total hip arthroplasty in octogenarians. A case-control study. J Bone Joint Surg Am. 2007; 89(12):2612-2618.
- Mahomed NN, Barrett J, Katz JN, Baron JA, Wright J, Losina E. Epidemiology of total knee replacement in the United States Medicare population. J Bone Joint Surg Am. 2005; 87(6):1222-1228.
- Mahomed NN, Barrett JA, Katz JN, et al. Rates and outcomes of primary and revision total hip replacement in the United States medicare population. J Bone Joint Surg Am. 2003; 85-A(1):2732.
- Katz JN, Losina E, Barrett J, et al. Association between hospital and surgeon procedure volume and outcomes of total hip replacement in the United States Medicare population. J Bone Joint Surg Am. 2001; 83-A(11):1622-1629.
- Hervey SL, Purves HR, Guller U, Toth AP, Vail TP, Pietrobon R. Provider volume of total knee arthroplasties and patient outcomes in the HCUP-Nationwide Inpatient Sample. J Bone Joint Surg Am. 2003; 85-A(9):1775-1783.
- Brauer CA, Rosen AB, Olchanski NV, Neumann PJ. Cost-utility analyses in orthopaedic surgery. J Bone Joint Surg Am. 2005; 87(6):1253-1259.
- Norman-Taylor FH, Palmer CR, Villar RN. Quality-of-life improvement compared after hip and knee replacement. J Bone Joint Surg Br. 1996; 78(1):74-77.
- Chang RW, Pellisier JM, Hazen GB. A cost-effectiveness analysis of total hip arthroplasty for osteoarthritis of the hip. JAMA. 1996; 275(11):858-865.
- Boettcher WG. Total hip arthroplasties in the elderly, morbidity, mortality, and cost effectiveness. Clin Orthop Relat Res. 1992; (274):30-34.
Drs Fehring (Thomas), Springer, and Griffin are from OrthoCarolina Hip and Knee Center; Mss Odum and Dennos are from OrthoCarolina Research Institute, Charlotte, North Carolina; and Dr Fehring (Keith) is from Virginia Commonwealth University Medical Center, Richmond, Virginia.
Drs Fehring (Thomas) and Griffin are both consultants for and receive royalties from Depuy Orthopedics, Inc. Mss Odum and Dennos and Dr Fehring (Keith) have no relevant financial relationships to disclose. Dr Springer is a consultant for and receives research support from Stryker Orthopedics.
Correspondence should be addressed to: Thomas K. Fehring, MD, OrthoCarolina, 2001 Vail Ave, Ste 200-A, Charlotte, NC 28207 (thomas.fehring@ orthocarolina.com).