The most common classification of periprosthetic femoral fractures is the Vancouver classification. The classification has been validated by multiple centers. Fractures are distinguished by location, stability of the femoral component, and bone quality. Although postoperative and intraoperative fractures are classified using the same three regions, the treatment algorithm is slightly different.

Type A fractures involve the greater and lesser trochanter. Fractures around the stem or just distal to the stem are Type B and subcategorised depending on stem stability and bone quality. Type C fractures are well distal to the stem and are treated independent of the stem with standard fixation techniques. The majority of fractures are either B1 (stable stem) or B2 (unstable stem). The stem is retained and ORIF of the fracture performed for B1 fractures. B2 and B3 fractures require stem revision with primary stem fixation distal to the fracture.

Intraoperative fractures use the same A, B, C regions but are subtyped 1–3 as cortical perforations, nondisplaced, and displaced unstable fractures, respectively. With the exception of A1 intraoperative fractures all other intraoperative fractures require surgical treatment.

A recent publication utilizing a New York state registry highlighted the patient risk of mortality associated with periprosthetic hip fractures. One month, 6 month and 1 year mortality was 3.2%, 3.8% and 9.7%, respectively. The mortality risk was lower for periprosthetic fractures treated with ORIF at 1 and 6 months compared to fractures requiring revision total hip.

The technique for removal of bone ingrown extensively coated devices involves cutting the stem below the metaphyseal portion of the stem, followed by removal of the proximal stem and trephine removal of the cylindrical distal portion of the stem. This can be done with or without an extended trochanteric osteotomy (ETO). When the proximal portion of the stem is not bone ingrown (extensive proximal osteolysis, or the stem is broken) or the metaphyseal bone is easily accessed (there is no collar) the stem can be cut through a bone window. In all other cases an ETO at the level where the stem becomes a cylinder is required to disrupt the metaphyseal bone prosthesis interface, cut the stem and extract the proximal portion of the stem.

Glassman described the techniques for removal of cementless stems in 1992. Forty-two loose stems were easily removed, 11 fibrous stable implants were removed with thin osteotomes, and 11 bone ingrown, canal filling, extensively coated stems were removed with trephines. In no cases was reconstruction precluded by stem removal. The critical tools required included manufacturer specific removal tools, high speed burs, thin osteotomes, universal extraction device for connection to the neck, and multiple trephines.

More recently, Kancherla reported the use of trephines to remove 36 porous coated stems. Eighty-six percent of cases were bone ingrown after removal, however complications included an extruded trephine causing a femoral fracture and two periprosthetic fractures thought to be secondary to trephine induced osteonecrosis. The authors recommend bypassing the most distally trephined bone by a minimum of 4cm.

Trephines are very helpful for removing distally fixed stems. Multiple trephines need to be irrigated and changed frequently to avoid dull cutting teeth which can lead to bone necrosis.

Ceramic-on-polyethylene (COP) bearings have traditionally been reserved for younger patients that were at high risk of polyethylene wear requiring revision. With the 1999 advent of highly crosslinked polyethylene (XLP), wear with XLP has not been a cause for revision. Simulator studies have not shown a difference in wear comparing COP to metal-on-polyethylene (MOP). Therefore, and considering the additional cost of COP, we have until recently not needed COP. However, a 2012 report of 10 cases that developed an adverse reaction to metal debris generated by head neck corrosion has resulted in COP becoming the most common bearing surface as reported by the American Joint Replacement Registry. This reactionary change has occurred despite the fact that we do not understand the cause, do not know the frequency, if it is more common in some implants than others, and we do not know the additional cost or markup of ceramic heads. One study reported a 3.2% revision prevalence caused by mechanically assisted crevice corrosion (MACC) at the head neck junction of a single manufacturer's implant. Other studies have estimated the frequency to be less than 5%. COST IS THE CONCERN in a value based healthcare environment. Models for and against the wholesale use of COP have been proposed and are based on variables that are unknown, including estimated frequency of the problem and the incrementally higher cost of a ceramic head. I use COP in younger patients that I believe will use their hip for more than 15 years. This is based on my personal experience. I have prospectively followed a series of MOP patients for 5 years and not seen cobalt elevations. I have placed new metal femoral heads on corroded femoral tapers without subsequent failure. I have evaluated the taper junctions of postmortem retrievals and found them virtually free of corrosion. A query of our institutional database for MOP primary hips identified 3012 cases between 2006–2017. Eighty revisions (2.7%) were identified. 2 of the 80 were for MACC representing 2.5% of revisions done on our own patients and 0.07% of our MOP cases. Further, evaluating our most recent all cause 350 revisions (7/2015–10/2017) there were 3 revisions for MACC (0.9%). Each one of us needs to EVALUATE OUR OWN PRACTICE AND MAKE AN EDUCATED, VALUE BASED DECISION whether or not to use COP in all patients.

Success in knee revision begins in the office. The initial evaluations determine the implant design and pre-operative diagnosis. The physical examination identifies the presence of instability, stiffness, extensor mechanism malfunction and previous incisions all of which influence the planned procedure. Prior to surgery arrangements are made to have all manner of revision implants, removal tools, and allograft material available.

Removal of implants must be done with a focus on preserving bone stock and the extensor mechanism. Initial exposure involves release of the gutters, lateral subluxation of the patella and removal of the polyethylene insert. These maneuvers combined with a quadriceps snip provide exposure for implant removal in 80–90% of cases. More extensive exposure options include quadriceps turndown, tibial tubercle osteotomy, medial epicondylar osteotomy and a femoral peel.

Tools needed for implant removal include thin osteotomes, offset osteotomes, thin saws and a high-speed bur. After polyethylene removal the femur followed by the tibia are removed. In many cases the existing well-fixed patellar component can remain. The implant cement or implant bone interface is approached for cemented and cementless implants, respectively. Tools are always directed parallel to the fixation surface. Offset osteotomes are helpful gaining access to the femoral notch when femoral pegs prevent access from the sides. Central keels or peripheral pegs can complicate tibial removal. Working completely around the keel from medial and lateral disrupts the peripheral tibial interface leaving just the central posterior metaphysis. Stacked osteotomes or a slap hammer can be used to lift the baseplate from the tibia.

Greater trochanter fractures after total hip replacement have been reported in up to 5% of cases. The outcomes are generally poor. Treatment options include non-operative care or surgical treatment with cerclage wires or a claw plate. We present a simple tension band technique for acute fractures with a single bony fracture fragment. We have not used the technique for chronic or comminuted fractures.

Technique: 2.5mm k-wires are passed through the fragment and anterior and posterior to the femoral implant. Eighteen-gauge wire is passed through a drill hole in the femur distal to the fracture and around the k-wires in a figure eight. The patient is kept 50% weightbearing with no active abduction for 4 weeks.

In four cases the fracture has gone on to healing. Patients have had a negative Trendelenburg sign without peritrochanteric pain.

The tension band technique is familiar to surgeons and has been reliable.

Porous-coated acetabular hemispherical components have proven successful in all but the most severe revision acetabular defects. A revision jumbo porous coated component has been defined as a cup with minimum diameter of 66mm in men and 62mm in women. In published studies this size cup is used in 14–39% of acetabular revisions. The advantages of this technique are ease of use, most deficiencies can be treated without structural graft, host bone contact with the porous surface is maximised, and the hip center is generally normal. Jumbo cups are typically used in Paprosky Type 2, 3A, and many 3B defects. Requirements for success include circumferential acetabular exposure, an intact posterior column, and much of the posterior wall. The cup should be stable with a press-fit between the ischium and anterior superior acetabulum with the addition of some superior lateral support. Additional support is provided with multiple dome or rim screws. Survivorship of the metal shell with revision for any reason has been reported to be 80%-96% at time frames from 15–20 years. The most common post-operative complication is dislocation.

Ceramic-on-polyethylene (COP) bearings have traditionally been reserved for younger patients that were at high risk of polyethylene wear requiring revision. With the 1999 advent of highly crosslinked polyethylene (XLP), wear with XLP has not been a cause for revision. Simulator studies have not shown a difference in wear comparing COP to metal-on-polyethylene (MOP). Therefore, and considering the additional cost of COP, we have until recently not needed COP. However, a 2012 report of 10 cases that developed an adverse reaction to metal debris generated by head neck corrosion has resulted in COP becoming the most common bearing surface as reported by the American Joint Replacement Registry. This reactionary change has occurred despite the fact that we do not understand the cause, do not know the frequency, if it is more common in some implants than others, and we do not know the additional cost or markup of ceramic heads. One study reported a 3.2% revision prevalence caused by mechanically assisted crevice corrosion (MACC) at the head neck junction of a single manufacturer's implant. Other studies have estimated the frequency to be less than 5%. COST IS THE CONCERN in a value based health care environment. Models for and against the wholesale use of COP have been proposed and are based on variables that are unknown, including estimated frequency of the problem and the incremental cost of a ceramic head. I use COP in younger patients that I believe will use their hip for more than 15 years. This is based on my personal experience. I have prospectively followed a series of MOP patients for 5 years and not seen cobalt elevations. I have placed new metal femoral heads on corroded femoral tapers without subsequent failure. I have evaluated the taper junctions of postmortem retrievals and found them virtually free of corrosion. A query of our institutional database for MOP primary hips identified 3012 cases between 2006–2017. Eighty revisions (2.7%) were identified. Two of the 80 were for MACC representing 2.5% of revisions done on our own patients and 0.07% of our MOP cases. Further, evaluating our most recent all cause 350 revisions (7/2015-10/2017) there were 3 revisions for MACC (0.9%). Each one of us needs to EVALUATE OUR OWN PRACTICE AND MAKE AN EDUCATED, VALUE BASED DECISION whether or not to use COP in all patients.

Porous-coated acetabular hemispherical components have proven successful in all but the most severe revision acetabular defects. A revision jumbo porous coated component has been defined as cup with minimum diameter of 66 mm in men and 62 mm in women. In published studies this size cup is used in 14–39% of acetabular revisions. The advantages of this technique are ease of use, most deficiencies can be treated without structural graft, host bone contact with the porous surface is maximised, and the hip center is generally normal. Jumbo cups are typically used in Paprosky type 2, 3A, and many 3B defects. Requirements for success include circumferential acetabular exposure, an intact posterior column, and much of the posterior wall. The cup should be stable with a press-fit between the ischium and anterior superior acetabulum with the addition of some superior lateral support. Additional support is provided with multiple dome or rim screws. Survivorship of the metal shell with revision for any reason has been reported to be 80%-96% at time frames from 15–20 years. The most common post-operative complication is dislocation.

Extensor mechanism complications after or during total knee arthroplasty are problematic. The prevalence ranges from 1–12% in TKR patients. Treatment results for these problems are inferior to the results of similar problems in non-TKR patients. Furthermore, the treatment algorithm is fundamentally different from that of non-TKR patients. The surgeon's first question does not focus on primary fixation; rather the surgeon must ask if the patient needs surgery and if so am I prepared to augment the repair? Quadriceps tendon rupture, periprosthetic patellar fracture, and patellar tendon rupture have similar treatment algorithms. Patients who are able to perform a straight leg raise and have less than a 20-degree extensor lag are generally treated non-operatively with extension bracing. The remaining patients will need surgical reconstruction of the extensor mechanism. Loose patellar components are removed. Primary repair alone is associated with poor results. Whole extensor mechanism allograft, Achilles tendon allograft, and synthetic mesh reconstruction are the current techniques for augmentation. In the acute setting if these are not available hamstring tendon harvest and augmentation is an option. Achilles tendons and synthetic mesh are easier to obtain than and entire extensor mechanism but are limited to patients that have an intact patella and the patella that can be mobilised to within 2–3 cm of the joint line. No matter which technique is used the principles are: rigid distal/tubercle fixation, coverage of allograft/mesh with host tissue to decrease infection, tensioning the augment material in extension, no flexion testing of reconstruction and post-operative extension bracing.

Do we need new polyethylene? Is there a clinical problem with first generation crosslinked polyethylene (XLPE)? Are we being duped into believing that doped polyethylene will solve a problem?

Clinical failures of polyethylene bearing total hip replacements are related to wear and the mechanical properties of the polyethylene. Wear is primarily related to crosslinking. Wear failures are secondary to periprosthetic osteolysis while mechanical failure causes cracking of thin polyethylene. Use of large femoral heads that reduce dislocation may increase wear and mechanical failure in the second decade of XLPE use.

There is no question that XLPE has reduced 2-dimensional (2D) head penetration, volumetric penetration and periprosthetic osteolysis with traditional 28 mm head sizes. Reported 2D penetration rates are 0.03–0.07 mm/year and clinically important polyethylene wear induced osteolysis is nonexistent. However, larger heads with the same 2D head penetration will generate more volumetric debris and could cause osteolysis.

There is no question that retrieved XLPE components have low levels of oxidation at the time of explant. While this is unexpected, the levels are well below levels reported with traditional polyethylene. It remains to be seen if these levels of oxidation will cause mechanical failures.

Currently available versions of polyethylene have focused on eliminating oxidation induced mechanical property reduction and not additional wear reduction. This is accomplished with Vitamin E doping or blending. While the local effects of Vitamin E polyethylene particles at the cellular level have been studied the clinical effect of these chemically new particles remains to be seen.

This author believes that long term volumetric wear with large head size is a greater concern than reduced mechanical properties secondary to in-vivo oxidation. New polyethylene development needs to focus on additional wear reduction. Can we afford to pay more for a new polyethylene in a value based healthcare environment?

INTRODUCTION

Adverse local tissue reactions (ALTR) and elevated serum metal ion levels secondary to fretting and corrosion at head-neck junctions in modular total hip arthroplasty (THA) designs have raised concern in recent years. Factors implicated in these processes include trunnion geometry, head-trunnion material couple, femoral head diameter, head length, force of head impaction at the time of surgery, and length of implantation. Our understanding of fretting and corrosion in vivo is based largely on the analysis of retrieved prostheses explanted for reasons related to clinical failure. Little is known about the natural history of head-neck tapers in well-functioning total hip replacements. We identified ten well-functioning THA prostheses retrieved at autopsy. We sought to determine the pull-off strength required for disassembly and to characterize fretting and corrosion apparent at the head-neck junctions of THAs that had been functioning appropriately in vivo.

Abductor deficiency commonly contributes to total hip dislocation. Successful treatment of the deficiency can improve function, decrease pain, and decrease reliance on implants to cure recurrent dislocation. The defining physical exam findings are dependence on ambulatory assistive devices, severe limp, positive Trendelenberg sign, and inability to abduct against gravity.

Three techniques have been described for chronic abductor discontinuity in which the abductors have retracted or are absent and cannot reach the greater trochanter: Vastus lateralis muscle shift, Achilles tendon allograft, and gluteus maximus muscle transfer. None of the techniques were specifically performed for dislocation.

The vastus lateralis shift transfers the entire muscle proximally maintaining the neurovascular bundle. The procedure requires an incision from the hip to the knee, isolation of the neurovascular bundle, and elevation of the muscle from the femur. The authors admitted that the technique is demanding and not easily applicable to many surgeons.

Repair with an Achilles allograft requires an identifiable contractile abductor mass. The allograft is looped through the abductors to bridge the gap to the trochanter.

Two variations of a gluteus muscle transfer for abductor deficiency after total hip have been described. A portion of the gluteus maximus with its distal fascial portion are transferred to the greater trochanter. As far as dislocation is concerned an advantage of this technique is the use of the posterior maximus flap to fill a posterior and superior capsular defect not addressed with the other techniques. In addition the technique is easy to perform in almost all cases.

The custom triflange acetabular component has been advocated for severe acetabular defects and pelvic discontinuity, cases in which a porous-coated hemisphere will not work. These are AAOS type III or IV defects, or alternatively classified as Paprosky 3B. Many have a pelvic discontinuity. A preoperative CT of the pelvis is sent to the manufacturer who generates a one to one scale 3D model of the hemipelvis. If the visualised defect cannot be treated with traditional methods then a triflanged component is created. Initial rigid fixation is obtained with screw fixation to the ilium and ischium. Subsequent bone ingrowth can provide long term fixation. The goal is to span the acetabular defect and obtain fixation to ilium and ischium with a third flange which rests on the pubis. Christie first reported on 67 hips (half with a discontinuity) with a mean follow-up of 53 months. No components were removed. There was an 8% reoperation for dislocation, 6% partial sciatic nerve palsy. Dennis reported 26 hips with a mean 54 month follow-up. Eighty-eight percent were considered successful. Taunton reported 57 cases with a pelvic discontinuity treated with a triflange at mean follow-up of 65 months. Eighty-one percent had a stable component and a healed pelvic discontinuity. The primary disadvantage of the technique is the preoperative time required to manufacture the device – typically 4–8 weeks.

We maintain a database on 1000 femoral revisions using extensively porous-coated stems. Using femoral rerevision for any reason as an endpoint, the survivorship is 99 ± 0.8% (95% confidence interval) at 2 years, 97 ± 1.3% at 5 years, 95.6 ± 1.8% at 10 years, and 94.5 ± 2.2% at 15 years. Similar to Moreland and Paprosky, we have identified prerevision bone stock as a factor affecting femoral fixation. When the cortical damage involved bone more than 10 cm below the lesser trochanter, the survivorship, using femoral rerevision for any reason or definite radiographic loosening as an endpoint, was reduced significantly, as compared with femoral revisions with less cortical damage.

In addition to patients with Paprosky type 3B and 4 femoral defects there are rare patients with femoral canals smaller than 13.5 mm or larger than 26 mm that are not well suited to this technique. Eight and 10 inch stems 13.5 mm or smaller should be used with caution if there is no proximal bone support for fear of breaking. Patients with canals larger than 18 mm may be better suited for a titanium tapered stem with flutes. While a monolithic stem is slightly more difficult for a surgeon to insert than a modular femoral stem there is little worry about taper junction failure.

Success in knee revision begins in the office. The initial evaluations determine the implant design and pre-operative diagnosis. The physical examination identifies the presence of instability, stiffness, extensor mechanism malfunction and previous incisions all of which influence the planned procedure. Prior to surgery, arrangements are made to have all manner of revision implants, removal tools, and allograft material available.

Removal of implants must be done with a focus on preserving bone stock and the extensor mechanism. Initial exposure involves release of the gutters, lateral subluxation of the patella and removal of the polyethylene insert. These maneuvers combined with a quadriceps snip provide exposure for implant removal in 80–90% of cases. More extensive exposure options include quadriceps turndown, tibial tubercle osteotomy, medial epicondylar osteotomy and a femoral peel.

Tools needed for implant removal include thin osteotomes, offset osteotomes, thin saws and a high-speed bur. After polyethylene removal the femur followed by the tibia are removed. In many cases the existing well-fixed patellar component can remain. The implant cement or implant bone interface is approached for cemented and cementless implants, respectively. Tools are always directed parallel to the fixation surface. Offset osteotomes are helpful gaining access to the femoral notch when femoral pegs prevent access from the sides. Central keels or peripheral pegs can complicate tibial removal. Working completely around the keel from medial and lateral disrupts the peripheral tibial interface leaving just the central posterior metaphysis. Stacked osteotomes or a slap hammer can be used to lift the baseplate from the tibia.

Porous-coated acetabular hemispherical components have proven successful in all but the most severe revision acetabular defects. A revision jumbo porous coated component has been defined as a cup with minimum diameter of 66 mm in men and 62 mm in women. In published studies this size cup is used in 14% – 39% of acetabular revisions. The advantages of this technique are ease of use, most deficiencies can be treated without structural graft, host bone contact with the porous surface is maximised, and the hip center is generally normal. Jumbo cups are typically used in Paprosky type 2, 3A, and many 3B defects. Requirements for success include circumferential acetabular exposure, an intact posterior column, and much of the posterior wall. The cup should be stable with a press-fit between the ischium and anterior superior acetabulum with the addition of some superior lateral support. Additional support is provided with multiple dome or rim screws. Survivorship of the metal shell with revision for any reason has been reported to be 80% – 96% at time frames from 15 – 20 years. The most common post-operative complication is dislocation.

Extensor mechanism complications after or during total knee arthroplasty (TKA) are problematic. The prevalence ranges from 1%-12% in TKA patients. Treatment results for these problems are inferior to the results of similar problems in non-TKA patients. Furthermore, the treatment algorithm is fundamentally different from that of non-TKA patients. The surgeon's first question does not focus on primary fixation; rather the surgeon must ask if the patient needs surgery and if so am I prepared to augment the repair? Quadriceps tendon rupture, peri-prosthetic patellar fracture, and patellar tendon rupture have similar treatment algorithms. Patients who are able to perform a straight leg raise and have less than a 20-degree extensor lag are generally treated non-operatively with extension bracing. The remaining patients will need surgical reconstruction of the extensor mechanism. Loose patellar components are removed. Primary repair alone is associated with poor results. Whole extensor mechanism allograft, Achilles tendon allograft, and synthetic mesh reconstruction are the current techniques for augmentation. In the acute setting if these are not available, hamstring tendon harvest and augmentation is an option. Achilles tendons and synthetic mesh are easier to obtain than an entire extensor mechanism but are limited to patients that have an intact patella and the patella that can be mobilised to within 2–3 cm of the joint line. No matter which technique is used the principles are: rigid distal/tubercle fixation, coverage of allograft/mesh with host tissue to decrease infection, tensioning the augment material in extension, no flexion testing of reconstruction and post-operative extension bracing.

I use monolithic, cylindrical, fully porous coated femoral components for many femoral revisions. Our institutional database holds information on 1000 femoral revisions using extensively porous-coated stems. To date, 27 stems have been re-revised (14 for loosening, 4 for infection, 7 for stem fracture, 2 at time of periprosthetic femoral fracture). Using femoral re-revision for any reason as an end point, the survivorship is 99 ± 0.8% (95% confidence interval) at 2 years, 97 ± 1.3% at 5 years, 95.6 ± 1.8% at 10 years, and 94.5 ± 2.2% at 15 years.

Similar to Moreland and Paprosky, we have identified pre-revision bone stock as a factor affecting femoral fixation. Among the 777 femoral revisions graded for femoral bone loss, 59% of the femurs were graded as having no cortical damage before the revision, 29% had cortical damage extending no more than 10 cm below the lesser trochanter, and 12% had cortical damage that extended more than 10 cm below the lesser trochanter. When the cortical damage involved bone more than 10 cm below the lesser trochanter, the survivorship, using femoral re-revision for any reason or definite radiographic loosening as an end point, was reduced significantly, as compared with femoral revisions with less cortical damage.

In addition to patients with Paprosky type 3B and 4 femoral defects there are rare patients with femoral canals smaller than 13.5 mm or larger than 26 mm that are not well suited to this technique. Eight and 10” stems 13.5 or smaller should be used with caution if there is no proximal bone support for fear of breaking. Patients with canals larger than 18 mm may be better suited for a titanium tapered stem with flutes. While a monolithic stem is slightly more difficult for a surgeon to insert than a modular femoral stem there is little worry about taper junction failure.

Revision hip approaches can be divided into posterior, anterior, transgluteal, and transtrochanteric. The approach chosen is dictated by what needs to be exposed and the approaches with which the surgeon is comfortable.

The posterior approach remains posterior to the gluteus medius and protects the hip abductors. The disadvantage of a posterior approach is post-operative dislocation.

The direct anterior approach is currently enjoying popularity as a primary technique. Surgeons experienced in the primary technique are applying it to revision surgery. The anterior approaches also protect the hip abductors. The disadvantage is poor access to the posterior acetabular column and mobilization of the femur to gain access to the femoral diaphysis.

Transgluteal approaches split the gluteus medius typically keeping the anterior portion of the medius intact with the vastus lateralis. Proximal exposure is limited by the superior gluteal nerve, which is 4 cm above the tip of the trochanter. The disadvantage of the transgluteal approach is difficult access to the posterior acetabular column and occasional abductor weakness.

The advantage of both the anterior and transgluteal approaches is a lower dislocation rate.

All three approaches are acceptable for revisions that only require acetabular rim and proximal femoral exposure. More extensive exposure requires modifications to these approaches or the use of a transtrochanteric approach.

Transtrochanteric approaches are defined by the length of the osteotomy (conventional or extended) and if the vastus lateralis remains attached to the trochanteric fragment (slide).

Distally extended osteotomies improve access to the femur.

Osteotomies without a distal attachment to the lateralis can be retracted proximally thus improving exposure of the ilium.

The technique for removal of bone ingrown extensively coated devices involves cutting the stem below the metaphyseal portion of the stem, followed by removal of the proximal stem and trephine removal of the cylindrical distal portion of the stem. This can be done with or without an extended trochanteric osteotomy (ETO). When the proximal portion of the stem is not bone ingrown (extensive proximal osteolysis, or the stem is broken) or the metaphyseal bone is easily accessed (there is no collar) the stem can be cut through a bone window. In all other cases an ETO at the level where the stem becomes a cylinder is required to disrupt the metaphyseal bone prosthesis interface, cut the stem and extract the proximal portion of the stem.

Glassman described the techniques for removal of cementless stems in 1992. Forty-two loose stems were easily removed, 11 fibrous stable implants were removed with thin osteotomes, and 11 bone ingrown, canal filling, extensively coated stems were removed with trephines. In no cases was reconstruction precluded by stem removal. The critical tools required included manufacturer specific removal tools, high speed burs, thin osteotomes, universal extraction device for connection to the neck, and multiple trephines.

More recently, Kancherla reported the use of trephines to remove 36 porous coated stems. Eighty-six percent of cases were bone ingrown after removal, however, complications included an extruded trephine causing a femoral fracture and two periprosthetic fractures thought to be secondary to trephine induced osteonecrosis. The authors recommend bypassing the most distally trephined bone by a minimum of 4 cm.

Trephines are very helpful for removing distally fixed stems. Multiple trephines need to be irrigated and changed frequently to avoid dull cutting teeth which can lead to bone necrosis.