Epidemiology of Revision Total Hip Arthroplasty: An Indian Experience

Abstract

Background

With increasing numbers of primary total hip replacement (THR), there has been a substantial increase in revision total hip replacement (RTHR) surgeries. RTHR are complex joint reconstruction surgeries involving significant cost, expertise and infrastructure. With its significant socioeconomic impact, we need to keep a close watch on the epidemiological trends of these procedures.

Methods

We prospectively studied the first-time RTHR performed at our institution for a 7-year period (2011–2017). We looked at patient demographics, the workload of RTHR and its etiology. We reviewed the microbiological profiles of septic revisions.

Results

Of the 1244 THR procedures performed, 260 (21%) were first-time revisions. The predominant cause of revisions was a prosthetic infection (38%) followed by aseptic loosening (33%), instability (15%), peri-prosthetic fracture (11%) and implant breakage (3%). In the aseptic loosening group, 55% of cases had primary cemented implant, 44% had only stem loosening, 31% had cup loosening and 25% had both cup and stem loosening. In the early, midterm, and late-failure groups, prosthetic infection remained the main cause of failure. In 60% of the septic revisions, the offending organisms could not be identified and of those identified most (77%) were gram negative.

Conclusion

In our study, the RTHR burden was 21%, which is similar to historic revision data from the west (1998–2001) and twice as compared to recent trends from the west (9–11%). Unlike western data, which show aseptic loosening (30–60%) as the predominant cause of hip revisions, in our study infection was the number one cause (38%).

Introduction

Over the past two decades, developing countries like India have seen a rapid increase in joint replacement surgeries, which is akin to the international trend [1]. With an increase in primary joint replacement surgeries, there has been a perceptible rise in the number of revision joint replacement surgeries throughout the world [2]. Modern implant design and improved surgical techniques should result in improved long-term survival of adult joint reconstructions [3]. However, an absolute increase in the number of these procedures and its use in younger, active population, coupled with increasing longevity, has maintained a significant revision burden [4]. Spread of the procedure to smaller surgical centers with limited infrastructure and surgical expertise is also adding to the revision workload [5]. With their robust nationwide administrative health data, western countries have been able to study the pattern and causes of revision arthroplasty in their population to help in planning future healthcare needs [1, 2, 6–12]. Developing countries like India lack national or regional joint registries and data is available onlu from select institutional databases. A significant proportion of revision surgeries are still performed at state-sponsored facilities. Our center is a similar specialized high-volume joint replacement facility, which receives patients from a number of hospitals from all over India. Being a military facility, the center has top-end infrastructure, implants and expertise to manage these cases. The centre maintains an institutional joint registry. This prospective study looked at the epidemiology of revision hip replacements. We studied the causes of failure, patient demographics and time trend over 7 years (2011–2017). We also reviewed the microbiological profile for septic hip revisions.

Materials and Methods

Design

It is a descriptive study conducted at our joint replacement centre which is part of a tertiary care military research and referral hospital receiving cases from all over the country.

Data Source

All the patients reporting to our centre from January 1, 2011, till December 31, 2017 with a failed total and bipolar hip arthroplasty requiring first-time revision total hip replacement (RTHR) were enrolled in the study. The research fellow recorded the case details using a pre-structured proforma for all cases and interacted with the surgical team (VK, BD, GM, CS and SK) to explicitly enter the possible causes for revision and categorised them into septic and aseptic revisions. Meticulous details of microbiological studies of hip aspirates and intraoperative tissue cultures were kept to study the organisms involved in septic RTHR.

Variables

Annual volume of total hip replacement (THR) surgeries and RTHR were calculated from the institutional registry. The annual revision burden was calculated by dividing the number of RTHR by the total number of THR and RTHR. Patient demographics, causes of failures, and time-to-failure from the index surgery were also recorded. Further, we recorded whether the index THR was performed at our institute or elsewhere. Musculoskeletal Infection Society (MSIS) criteria were used to label the case as septic revision [13]. Hip aspiration was carried out whenever indicated. The microbiology reports were reviewed to determine the type of offending organism and antibiotic sensitivity pattern. In this study we did not look at the details of management and outcomes of revision surgery.

Statistics

Statistical analysis was performed with Stata Version 12 (Stata Corp LP, Texas USA). We reviewed the patient demographic data including age, sex, BMI, comorbidities and American Society of Anaesthesiologist (ASA) score in the primary THR vs RTHR group in terms of mean, mode and standard deviation if it was a continuous variable and rate if it was a dichotomous data. We then compared the failures as per cause and time to failure (Early < 2 years from index procedure; midterm 2–10 years; and late > 10 years). We looked at the proportion of failures as per cause. We also analysed the trend over time (2011–2017).

Results

Twelve hundred and forty-four adult hip reconstructions (primary and revisions) were performed from 01 January 2011 to 31 December 2017, of which 260 (21%) were first-time revisions. The index hip arthroplastywas performed for primary osteoarthritis in 144 patients, for fractures of the hip in 48 cases, Avascular Necrosis of the hip in 40 cases, inflammatory arthritis in 27 cases and one was done for dysplastic hip. The annual revision burden varied between 19 and 23% (Fig. 1). Most of the RTHR (66%) underwent primary THR procedures at other hospitals. Most RTHR (111 of 260; 42%) were in the early failure group (revised within 2 years of index surgery), followed by midterm (89 of 260; 34%) and late failures (60 of 260; 24%) (Fig. 2). In 17% only acetabular cup was revised, in 32% only femoral stem was revised and in 48% both cup and femoral stem were changed. Only head and liner exchange was done in 3% cases. The average age of patients undergoing revisions was 59.4 (± 12.6; 27–92 years), as compared to primary THR 54.7 (± 16; 10–95 years). In both groups, 60% of patients were males. 10% of patients in primary THR group and 15% patients in RTHR group were a poor risk for surgery (ASA 3 and 4).

Fig. 1.

Revision total hip replacement burden and causes (2011–2017)

Fig. 2.

Causes of revision: temporal profile

Prosthetic Infection was the cause of failure in 38% (100 patients), followed by aseptic loosening 33% (85 patients), dislocations 15% (40 patients), peri-prosthetic fracture 10% (27 patients) and implant breakage 3% (eight patients)(Fig. 1). When we looked at the temporal profile (Fig. 2), infection remained the predominant cause of early (45%) and midterm failures (41%). In 51% of the septic revisions (diagnosed based on MSIS criteria), the offending organisms could not be cultured. Of the 49% positive cultures, 71% (35 cases) were gram-negative infections and 29% (14 cases) were gram-positive infections (Table 1). Most of the delayed and late infections were due to gram-negative organisms or were culture negative (Table 2).

Table 1.

Microbiology studies of infected total hip replacements (culture report and antibiotic sensitivity of organism grown)

Infected THR	100
No growth	51
Growth	49

Gram positive: 14
Organism	No.	Antibiotic sensitivity
Staphylococcus aureus	12	Linezolid, Vancomycin, Trimethoprim//sulfamethoxaxole, Rifampicin
Staphylococcus saprophyticus	1	Linezolid, Teicoplanin, Vancomycin, Trimethoprim/Sulfamethoxazole
Staphylococcus epidermidis	1	Gentamicin, tegecycline, nitrofurantoin, Rifampicin

Gram negative: 35
Organism	No.	Antibiotic sensitivity
Acinetobacter species	3	Ciprofloxacin, amikacin, levoflox
E. coli	3	Amikacin, Piperacillin, Meropenem
Enterobacter cloacae	8	Ciprofloxacin, Moxifloxacin, Tigecycline, Rifampicin
Klebsiella oxytoca	1	Ampicillin, Piperacillin/Tazobactam, Ceftriaxone, Cefepime, Imipenem, Amikacin, Ciprofloxacin
Pseudomonas aeruginosa	8	Imipenem, Meropenem, Amikacin, Gentamicin, Tobramycin, Ciprofloxacin, Moxifoloxacin
Pseudomonas luteola	1	Ertapenem, Meropenem, amikacin, Gentamicin, Tobramycin
Klebsiella pneumoniae	8	Tigecyline, Colistin, Trimethoprim/Sulfamethoxazole
Sphingomonas paucimobilis	1	Moxifloxacin, Tigecycline, Meropenem
Proteus mirabilis	2	Colistin

Table 2.

Temporal profile of infections in terms of early, delayed and late infections from 2011 to 2017, along with the details of microbiology report

Year	Early (< 3 months)	Delayed (3 months–2 years)	Late > 2 years
2011	3	4	6
2012	2	7	6
2013	0	10	4
2014	2	7	6
2015	2	8	6
2016	1	8	5
2017	0	4	9
Total	10	48	42
Microbiology report	No growth—05	No growth—23	No growth—23
	Gram positive—02	Gram positive—04	Gram positive—08
	Gram negative—02	Gram negative—21	Gram negative—11

Aseptic loosening was the cause of RTHR in 85 cases (33%), predominantly observed in midterm (32% of all midterm failure cases) and late-failures (38% of all late failure cases) (Fig. 2). Primary arthroplasty was cemented in 55% of cases of aseptic loosening. Failed bipolar was the cause of RTHR in 31% cases of aseptic loosening, of which 66% were uncemented. Isolated stem loosening was the cause of RTHR in 44% cases and in another 25% cases, both stem and cup were loose. Isolated cup loosening was seen in 31% cases (Fig. 3). Most of the cups, which needed revision, were cemented (34/48; 71%),. whereas an equal number of cemented and uncemented stems were revised for aseptic loosening (Fig. 3).

Fig. 3.

Revision total hip replacement for aseptic loosening: distribution

Dislocation was the cause of RTHR in 15% cases. Most of the RTHR for dislocations (29 out of 40; 73%) were in the early failure group (Fig. 2). They constituted 26% of early failures only after prosthetic infections (45%). However, some cases of RTHR following dislocations were also observed in the midterm and late failure groups (Fig. 2). During revision surgery, acetabular cup version was found inappropriate in 53% cases, femoral stem version was found faulty in 25% cases and femoral offsets were less than required in 20% cases. In one case there was abductor loss (Fig. 4).

Fig. 4.

Revision total hip replacement for dislocation: causes

Peri-prosthetic fractures were seen as the cause of RTHR in 10% [26] cases. Most of these fractures were following trauma and 50% were seen within 2 years of primary THR (Fig. 2). 90% were of Vancouver type B, of which type B2 were 37% (ten cases) and B3 52% (14 cases) (Fig. 5).

Fig. 5.

Revision total hip replacement for peri-prosthetic fractures: distribution by Vancouver type

Implant breakage was seen in eight cases where the femoral stem broke. All but one were smooth stainless steel femoral stems used with bipolar heads. In four cases the stem was cemented and three uncemented. One uncemented THR stem broke.

Discussion

With the overwhelming success of joint replacement surgery, its popularity has spread and both very young and elderly patients with complex joint destruction are being offered this surgery. It was always understood that with very good 15- 20-year results, as these patients enter their third decade of follow-up the need for revisions would go up. Unlike primary joint replacement procedure, revision surgery requires a high level of technical expertise, exhaustive implant inventory and sophisticated infrastructure. This entails significant socioeconomic impact on the countries world over. Western countries are already closely looking at the epidemiological pattern of the revision of joint replacement surgeries to be able to plan future health care needs. They are putting in extensive efforts to find out ways in which they can improve the quality of primary joint replacement surgeries to minimize and delay the need for revision surgeries. In developing countries with the unorganized health sector and lack of consolidated administrative healthcare data, it becomes impossible to study the pattern of disease load and conduct an epidemiological study at the national level. Until the governments of these countries make any efforts to collate national health data, epidemiological studies of this kind would only be possible at the institutional level. Our study is one such single-centre study. Being a state-sponsored facility with its heterogeneous caseload of the representative patient population received as transferred cases from all over the country it gives us a realistic picture of volume, cause and temporal profile of the RTHR being done. We have also compared our findings with the recent literature from the western countries (Table 3).

Table 3.

Review of published studies on epidemiology of revision total hip replacement

S. no.	Studies	Years of operation	Type of study	Cup revision	Cup and stem	Only stem	Head and insert	na	Early revisionsb (%)	Infection (%)	Aseptic loosening	Instability (%)	Fracture (%)	Implant breakage
Institutional studies
1	Prokopetz et al. [14]	2003–2012	Systematic review	NA	NA	NA	NA	65	NA	29	46	19	NA	NA
2	Bozic et al. [15]	2005–2010	Retrospective review	14	43	14	13	235,857	NA	15	33	22	6	10
3	Jafari et al. [16]	2000–2007	Retrospective review level III	NA	NA	NA	NA	1366/1272	NA	8	65	15	5	1.75
4	Delaunay et al. [17]	2010–2011	Prospective	NA	49	NA	NA	258,461	35	11	53	19	12	3
5	Gwam et al. [18]	2009–2013	Retrospective review	15	41	14	15	820	49	13	27	17	5	3.3
6	Wright et al. [19]	1995–2008	Retrospective review	NA	NA	NA	NA	3647	NA	5	76	38	9	NA
7	Bozic et al. [9]	2005–2006	Retrospective review level II	12.7	41.1	13.2	12.6	51,345	NA	15	31	23	6	9.9
8	Ulrich et al. [20]	1996–2004	Retrospective review	NA	NA	NA	NA	237	NA	16	68	17	6	2.1
9	Springer et al. [21]	1986–2005	Retrospective review	NA	NA	NA	NA	1036	NA	11	61	16	6	NA
10	Haynes [22]	2004–2014	Retrospective review	NA	NA	NA	NA	870	NA	14	53	21	5	3.6
11	AOAJRR Report [8]	2015	Retrospective review	32.3	28.8	17.7	11.1	17,082	NA	14	48	14	10	2
12	Our study	2011–2017	Prospective study	17	48	32	3	NA	NA	38	33	15	10	3
13	Wetters et al. [27]	2004–2010		27	19	15	28	1152/1211	14	15	56	18	5

^aTotal revision THA performed

^bRevisions done within 2 years of index procedure

In developing countries like India, unlike Total Knee Replacement, THR has been performed in good numbers for the past four decades. That explains the high annual revision burden of THR (> 20%) as compared to TKR, which is still at 4% in our institute [23]. In the western countries, the historic (1999–2002) revision rates for THR were 11–18% and for TKR 7–11%. However, their recent data have shown a significant decline in these rates: 7–8% for knees and 9–11% for hips [24]. This trend could be due to improvement in implant design, adoption of newer techniques and improved material (use of highly cross-linked polyethylene). As the modern implant and techniques have now been widely used in our country we may also see a downward trend in annual revision burden in coming times. Annual RTHR burden (average 21%) at our centre is similar to high rates (11–18%) historic (1999–2002) hip revision burden of the west [2]. It is almost twice as much as the current revision hip burden (9–11%) in the west [15]. In our study PJI, aseptic loosening, dislocations and peri-prosthetic fractures were the predominant cause of failure, as also seen in most of the western data (Table 3).

A significant proportion of revisions in our study were early (42%; within 2 years), which were largely due to prosthetic joint infections (PJI), which contributed the maximum number of cases to our revision burden. Overall, 38% of cases were revised for PJI, which is unlike that of western literature where aseptic loosening remains the number one cause. In most western studies PJI contributed only 11–15% (Table 3). In our study both in early and midterm failures, PJI remained the number one cause of failures. In late failures, aseptic loosening was the predominant cause followed by PJI. In more than half the cases (51%) the offending organism could not be isolated complicating the management of these cases. Most of the recent western literature shows successful isolation of infecting organism in PJI up to 70%. They have used implant sonification and specialized microbiological techniques including extended cultures on specialized media [25]. In our centre, we did not have access to these techniques and our hospital used traditional microbiological techniques, which could be responsible for lower culture detection rates in MSIS diagnosed PJI. Although poor culture positivity rates in our study impair comparability of our findings with western literature, unlike the existing understanding that most of the PJI are gram-positive staphylococcal infections; most culture positive cases in our study were gram-negative organisms (71%). Literature from the west shows most hip PJI were caused by staphylococcus aureus (13%) or coagulase-negative staphylococcus (30%), gram negative and anaerobes were seen in 9% and 7% cases, respectively [26]. Gram-negative infections are usually hospital-acquired infections caused due to a breach in sterility during the perioperative care. When we look at the temporal trend (2011–2017), we see a rise in delayed and late infections involving gram-negative organism and fastidious organisms difficult to isolate. Non-adherence to strict asepsis and poor quality of perioperative care at various small and intermediate care facilities functioning in India may be responsible for such a temporal trend. Looking at the significant contribution of PJI to revision workload, there is an urgent need to evolve strict perioperative protocols including preoperative optimization of patients to minimize the rates of PJI. We also need to quickly adopt the most recent microbiological tools to identify infecting organism to help in the effective management of these cases.

Aseptic loosening as a cause of RTHR was seen in 34% of cases in our study. If we look at the western literature (Table 3) most studies show aseptic loosening and osteolysis as the number one cause of hip revision ranging from 48 to 76%. This has already been attributed to poly wear seen in old design cemented THR which used standard polyethylene and not the highly cross-linked polyethylene now available free in the market. In our study, 55% cases (47/85) had late failures of the cemented stems. The uncemented stem failures in our study were predominantly seen in cases in which poorly designed smooth stainless steel stems akin to Austin Moore Prosthesis were used with bipolar heads. Since these stems had no opportunity to integrate they failed in active community ambulators. In 56% (48 /85) of aseptic failures, acetabular loosening was seen. of which 71% (34) were cemented cups,. which could be attributed to poor quality of older generation polyethylene.

Prosthetic dislocation was the third commonest cause for RTHR observed in our study (15%); as compared to this, the contemporary literature from the west (Table 3), shows that 14–38% of RTHR were due to prosthetic dislocations. Most of the cases of RTHR which were done for dislocations were seen in the early postoperative period (73%; 29 out of 40 cases), most were due to technical errors of implant placement with compromised alignment and offset restoration. During revision, more than half the cases needed cup revision due to inappropriate cup version (53%). In 45% of cases, femoral stem version and offset had to be changed. Only one case had abductor loss. We did not study the possible aetiology of the dislocations, which could have varied from surgical errors to patient compliance and even trauma. But as the third commonest cause for RTHR, we need to review our surgical techniques, choice of implant design and placement to minimize postoperative instability. Already techniques are being devised to combat this complication like the use of safer approaches (direct anterior approach to hip), use of robotics and navigation for placement of implants and introduction of large diameter heads and dual mobility acetabular cups to drastically reduce dislocation risks.

Peri-prosthetic fracture accounted for revision in 27 patients (10%); this is similar to the incidence of peri-prosthetic fractures noted in contemporary studies on RTHR (Table 3). In 16 cases it was seen early after use of an uncemented stem indicating surgical technique related errors and in ten cases it was around a loose-cemented stem seen in late failure group following osteolysis.

There were eight cases of stem breakage, seven of which were stainless steel polished poorly designed stems used with cement in 04 cases and without cement in three cases along with a bipolar acetabular component. Possible cause of metal fatigue and failure in these cases was poor metallurgy and design of these stems along with use in active community ambulators. Similar rates of implant breakage (3–10%) have been shown by recent studies from the west (Table 3).

Our study has a few limitations. Although we calculated the annual burden of revision, it was not possible to calculate the incidence of RTHR as we had many patients in our study who underwent primary THR at other centre and it was not possible to quantify baseline population. In the absence of a national registry, such data coming from a state-sponsored high-volume tertiary care facility catering to a homogeneous population are vital in planning healthcare needs of the community and interventions to improve arthroplasty care. We did not evaluate the management strategies including details of revision surgery and outcomes of these revisions in this study. With the existing database, we plan to evaluate mid- and long-term outcomes of these revisions in our future studies. A major strength of our study is that unlike national administrative healthcare data or registry data, our data collection, being prospective in design and analysed by a dedicated research team on a case-to-case basis, is likely to be highly accurate in defining the causes of revision.

The predominant cause of hip revisions in our centre over the past 7 years was Prosthetic Joint Infections. Due to inadequate microbiological support and use of empiric non-structured antibiotic regimen, in many of the septic revision, the offending organism is not found which complicates their management. Even in culture-positive cases, unlike the western countries, predominant organisms are a gram-negative hospital-acquired organisms, which makes treatment even more difficult. Hospital Acquired Infections (HAI) are preventable and we need to improve practices to control PJI resulting from HAI. As compared to TKR, THR was introduced in our country a decade or two earlier and with a larger baseline population of THR patients who have now lived more than two to three decades, we are already seeing a rising trend in aseptic loosening-related failures in RTHR patients. Hence, in coming times, we need to ensure the availability of expertise, infrastructure and economic resources to take care of the increasing demands of RTHR.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical standard statement

This article does not contain any studies with human or animal subjects performed by the any of the authors.