Bilateral simultaneous total knee replacement surgery remains controversial with arguments for and against its use. Doing sequentially staged TKR's is a safer procedure and may have additional benefits as set out below.

If both knees need to be replaced we have often seen that the symptoms of the contralateral knee improve after the one knee is replaced and that patients wait some time before having the opposite knee replaced.

At the 2010 Effort congress Prof Hernigou of France stated that you never need to template knee x-rays as there is an absolute association with patients height and implant size. Templating of the knee for size is seldom done in clinical practice but could be handy when doing revision surgery where normal anatomy has been lost. This is however difficult with digital x-rays due to enlargement problems.

With this in mind we retrospectively looked at the size of knee implants inserted to see if there was any relation with patient's height and also to see if this differs in male and female patients.

Introduction

Digital x-rays on computer screens are difficult to template due to the lack of standardized magnification. This can be overcome by the use of markers placed onto or next to the patient but have certain shortcomings. Trochanteric marker placements are operator dependant and very difficult to use in the obese patient. Inter- thigh markers are also operator dependent and often embarrassing for radiographer and patient. Anterior combined with posterior markers are very accurate (King et al) but can only be used with a digital template system which is costly and time consuming. We would like to describe a new method of posterior bar markers that are easy to use with standard hip templates.

1346 Primary TKR’s were evaluated. In keeping with the principle of Insall all patellas were resurfaced with the only exclusion being a previous patellectomy or excessive patella erosion.

Most TKR were of posterior cruciate substituting devices (IB11 (56.9%) or Nexgen LPS (42.3%)). The reason for operation was OA (94.5%), RA (2.9%), and others 2.6%. Most knees were in varus (68.5%), 17% were in valgus, and 14.5% were in neutral alignment.

The method of preparing the patella and extensor mechanism was as follows: A total fat pad excision was performed, debulking the patella thickness of 1mm. The patella component was placed medially and superiorly, a peri-patella synovectomy was performed, and a release of the lateral patella femoral ligaments was done. A lateral release was performed in 17.5% of patients.

Follow up ranges from 9 months to 15 years. Reoperation for patella problems was necessary in only 5 patients (0.37%). There was 1 case of patella subluxation, 1 case of persistent anterior knee pain, and 3 patients with a patella clunk (in IB 11 knees only)

In our hands this approach has led to excellent long term results without some of the potential complications described in the literature and warrants continued use of routine patella resurfacing when doing TKR.

Knee sepsis following TKR can have devastating consequences for patient as well as surgeon. A two stage revision is a well accepted technique in TKR sepsis with the introduction of a temporary antibiotic cement spacer being the most popular procedure although irrigation techniques are popular in SA.

From a total of 111 revisions TKR from my practice 26 (23%) were 2 stage revisions for joint sepsis following TKR. 3 cases were early, 10 intermediate and 13 late onset sepsis cases. Most common organism was S. Aureus (7/26) and S. Epidermidis (7/26) although numerous other organisms were seen.

In all cases a two stage revision with a Palacos R cements spacer plus parenteral antibiotics were used. Prosthesis used for revision was primary knee prosthesis in 8 cases and revision (stemmed) prosthesis in 18 cases. Follow up range from 13 years to 6 months (average 6.8 years) with only one case of recurrent sepsis (3.8%) which went on to an arthrodesis. Time from debridement and spacer placement to revision TKR varied from 3 weeks to 10 months (average 2.1 months).

This paper shows that meticulous debridement followed by standard antibiotic cement spacer technique with additional parenteral antibiotics is indeed the gold standard approach without necessitating additional irrigation techniques.

The number of worldwide THRs is growing but because we have no national register, the number done in South Africa is unknown. This is the third survey attempting to track the number of THRs done in this country. A survey 6 years ago indicated that 8986 THRs were done annually.

We sent out 521 questionnaires to members of the South African Orthopaedic Association. To date we have received 166 (31.86%) responses from members, 94 of whom perform THR. The data to date show that 4031 THRs are done annually, a mean of 42.88 operations per member.

Fully cemented THR is still the most popular form (58.3%), followed by hybrid (25.37%) and uncemented (15.85%). The most popular cements are Palacos (65%) and CMW (28%). Four prostheses lead the field at this stage: Elite Plus (27%), C-stem (9%), osteal (7%) and metal-on-metal resurfacing (6%).

We expect to have at least a 95% response by September 2004.

Hip pathology can exacerbate symptoms of low back pathology. In patients with both, it can be difficult to evaluate back symptoms and hip pathology can be missed. From our hip register, we found that 66 of 814 THA patients (8.11%) had also undergone spinal surgery, 92.4% before THA. Among these 66 patients were 15 (24% or 1.84% of the total number) in whom back surgery did not relieve pain. Their pain disappeared after THA. It may be significant that a neurosurgeon performed the spinal surgery in all cases.

We suggest that orthopaedic surgeons examining patients with back pain always look for a Trendelenburg gait, insist on anteroposterior radiographs of the pelvis, routinely examine the hip when examining the low back and, if dual pathology exists, consider doing THA first.

One of the problems of keeping an orthopaedic register is that radiographs, which are an integral part of records (particularly with total joint arthroplasties) are cumbersome and costly to store.

We have developed a cheap and simple digital storage system, photographing radiographs with a digital camera and storing pictures in a specially designed database. Retrieval is rapid and hard copies can be printed.

Over 10 000 arthroplasty radiographs have been digitalised by this inexpensive method, which could prove useful for most orthopaedic surgeons.

Previous incisions around the knee may complicate subsequent total knee arthroplasty (TKA) because they can lead to skin problems, with wound breakdown and a risk of sepsis.

Our database contains details of 925 TKAs, 851 primary and 74 revision procedures. Of the 851 primary TKA patients, 368 had previously undergone knee surgery, 72 of them more than once. Twenty of the 74 patients who underwent revision TKA had undergone one previous procedure (excluding the primary TKA), and 24 had undergone multiple procedures. We clinically reviewed 133 TKAs, classifying previous procedures into midline (24), medial (50), lateral (26) and transverse (13) procedures. In 53 cases there had been previous arthroscopic procedures. Excluding the arthroscopies, previous scars were followed in 20 cases, partially followed in 11 cases and ignored in 53 cases.

Following up patients for a minimum of six months, we saw only six cases with minor wound edge slough. These did not require further surgery. Three of the six patients were in the group of 442 with previous scars, and three in the group of 483 without previous scars. All patients had spinal anaesthesia, peri-operative oxygen, vacuum drainage and a delayed knee-bending program, which we believe contributed to the low incidence of wound problems.

We believe that previous scars should be followed if they are approximately in the line of a normal midline TKA incision, and that scars beyond the midline can be ignored without increasing the risk of skin necrosis.

Previous incisions around the knee may complicate subsequent total knee replacement (TKR) surgery because they can lead to skin problems, with wound breakdown and a risk of sepsis.

Our database contains details of 925 TKRs, 851 primary and 74 revision procedures. Of the 851 primary TKR patients, 368 had previously undergone knee surgery, 72 of them more than once. Twenty of the 74 patients who underwent revision TKR had undergone one previous procedure (excluding the primary TKR), and 24 had undergone multiple procedures. We clinically reviewed 133 TKRs, classifying previous procedures into midline (24), medial (50), lateral (26) and transverse (13) procedures. In 53 cases there had been previous arthroscopic procedures. Excluding the arthroscopies, previous scars were followed in 20 cases, partially followed in 11 cases and ignored in 53 cases.

Following up patients for a minimum of six months, we saw only six cases with minor wound edge slough. These did not require further surgery. Three of the six patients were in the group of 442 with previous scars, and three in the group of 483 without previous scars. All patients had spinal anaesthesia, peri-operative oxygen, vacuum drainage and a delayed knee-bending program, which we believe contributed to the low incidence of wound problems.

We believe that previous scars should be followed if they are approximately in the line of a normal midline TKR incision, and that scars beyond the midline can be ignored without increasing the risk of skin necrosis.

Revision total hip arthroplasty (THA) may be indicated for reasons other than femoral loosening.

From 1991 to 1999, 190 revision THA procedures were performed. These included 39 cement-on-cement (20.5%), 68 bone impaction (35.8%), 31 long stem cemented (16.3%), 16 acetabulum only (8.4%), six by-pass prosthesis (3.2%), 20 short stem cemented (10.5%) and 10 miscellaneous revisions (5.3%).

The mean time from previous THA was 6.6 years (1 to 23). Of the cement-on-cement revisions 18 (46%) were done for acetabular loosening, 13 (33.5%) for chronic dislocation, seven (18%) for fracture of the femoral prosthesis and one (2.5%) for chondrolysis of the hemiprosthesis. At a mean short follow-up of three years (1 to 7), we have seen no loosening of the femoral prosthesis.

The absolute indication for this procedure is a Type-A cement mantle in Gruen zones 2 to 6. Cement-on-cement revision can be done only in selected cases, but when possible this technique saves time and money and reduces the perioperative risk.

We present five case studies of a new technique for the treatment of distal femur fractures after total knee arthroplasty. This type of fracture is rare, but when it occurs can present a dilemma as to the correct treatment. If the prosthesis is loose, the logical treatment is revision surgery with the use of long stem stabilisation. If the prostheses are firmly fixed, the best method of treatment is difficult to determine. Intramedullary fixation is a well-known modality, but proper fixation distal to the fracture can be problematic in very distal fractures.

We performed intramedullary fixation of these fractures, using standard retrograde condylar locking nails inserted through the notch of the femoral prostheses. The problem of distal fixation was solved by fixating the nail to the femoral prostheses with a plate that fitted into the notch of the pros-theses and was securely fixed to the nail with a custom-made screw. This not only gave alignment stability but also aided in compression of the fracture. A locking screw distal to the fracture line was inserted in some patients to aid fixation. but could not be placed in others owing to the distal position of the fracture. Autograft was used in most cases to aid fracture healing. Postoperatively the leg was immobilised in a cast for six weeks.

This method of fixation of the nail to the prostheses has not been described in the literature to date. We believe that this technique offers a new modality in treating these complex fractures, providing adequate fixation, alignment and compression stability.