Introduction: Blood loss is a major concern following total knee arthroplasty (TKA) frequently resulting in blood transfusions postoperatively. Various strategies exist to reduce blood loss and allogenic transfusion requirements. This study investigates the effect of immediate postoperative flexion on blood loss and transfusion requirements following TKA.

Methods: 180 consented patients undergoing primary TKA by a single surgeon were enrolled into a prospective randomised controlled study. 90 patients were randomised to have the operated knee nursed in extension postoperatively, and 90 patients to have the knee nursed in flexion for six hours postoperatively. Both groups followed a strict transfusion protocol. Data collected included calculated pre- and postoperative haemoglobin and haematocrit which was used to calculate total blood loss. Units transfused and postoperative complication rates were also recorded.

Results: There was no significant difference in demographics or factors predisposing to bleeding between the groups. The mean total blood loss was 1841mls for those in the extension group compared with 1587mls in the flexion group (p=0.02). The mean number of units transfused in the extension group was 0.78 units/patient compared with 0.36 units/patient in the flexion group (p=0.004). There was no significant difference in pain scores between the groups (p= 0.62).

Conclusion: This study shows that the use of immediate postoperative flexion significantly reduces calculated total blood loss and transfusion rates following TKA.

Purpose: The objective of this study was to observe the change in transfusion rate by decreasing the transfusion trigger for patients undergoing primary total knee arthroplasty.

There are no transfusion guidelines defined for Total knee arthroplasty (TKA), and each orthopaedic unit usually has its own protocol when it comes to transfusion. With the present climate of concern about non-autologous blood transfusions and transmission of infection, orthopaedic units are trying to keep their transfusion rates to a minimum and only transfuse when absolutely necessary.

Method: The transfusion trigger prior to protocol change was a haematocrit (Hct) < 0.31, which was then changed to a lesser trigger of Hct < 0.25. We collected data on 331 patients; 181 patients prior to the change and 150 patients after the change. Each patient had undergone a primary TKA by a single surgeon using the same operative technique.

Results: In the group prior to change in transfusion trigger (Hct < 0.31), 76% of patients were transfused 88% of females and 54% of males. In this pre-trigger change group, 100% of patients with a pre-operative Hb of 12g/dl or less were transfused.

Of the 150 patients reviewed after the change in transfusion trigger (Hct < 0.25), only 29% were transfused – 35% of females and 18% of males. 50% of patients with a pre-operative Hb of 12 g/dl or less were transfused.

Conclusion: A change in the transfusion trigger greatly reduced the transfusion rate (reduction of 45%) with no adverse patient events. Obviously, this has both clinical and financial benefits. Pre-operative Hb is the best indicator as to whether a patient will require a transfusion following TKA. In order to decrease transfusion rates further, patients with a pre-operative Hb < 12 g/dl may benefit from pre-operative boosting of their Hb.

Purpose: The objective of this study was to observe the change in dislocation rate by introducing a capsular enhanced short external rotator repair during primary total hip arthroplasty.

Dislocation after primary total hip arthroplasty is a relatively common complication. The incidence is usually between 1% and 4%, although 0% to 10% has been reported in the literature. The posterior approach is by far the most popular by surgeons, but it reportedly has an increase risk of post-operative dislocation. The posterior approach causes disruption of the posterior capsule and the short external rotators. Studies have shown that repair of the posterior capsule and the short external rotators greatly reduce the dislocation rate.

Method: Review of the dislocation rate of THR’s by a single surgeon before and after the introduction of a capsular enhanced short external rotator repair (August 2002). It was a sequential cohort of 2213 Total Hip Replacements (THR), inserted over a period from January 1999 to June 2004. All cases were done by the senior author through a posterior approach using a Belfast custom stem, a 28mm head, and a cementless socket. We collected data on all dislocations which occurred on the hips inserted during this time period. We were therefore able to compare dislocation rates for THR’s done before and after the introduction of the soft tissue repair. The posterior capsule and short external rotators were reattached to the greater trochanter through drill holes using a number 5 Ethibond.

Results: This dislocation rate before the introduction of the new posterior soft tissue repair was 3.9% (58 out of 1501 THR’s). After the introduction of the repair the dislocation rate was 0.9% (6 out of 712 THR’s). 81% (47 of58) of dislocations before the change of practice were classed as early dislocations (with 3 months of surgery). 83% (5 of 6) were clearly dislocators after the new soft tissue repair. Considering only the early dislocations, the dislocation rate reduced from 3.1 to 0.7%.

Conclusion: Considering our reduction from 3.9% to0.9%, and results of previous studies, it is clear that the capsular enhanced and short external rotator repair is associated closely with a decreased dislocation rate. We feel other factors such as restoration of joint centre ensuring correct orientation of components are important in conjunction with the posterior soft tissue repair to keep dislocations to a minimum.

Purpose: The objective of this study was to assess whether a correlation existed between blood loss and soft tissue release carried out during total knee replacement in the valgus knee.

Method: We identified 208 patients with a valgus deformity undergoing primary total knee arthroplasty . These were performed over a 33 month period by a single surgeon. An LCS mobile bearing prosthesis was used in all patients. The valgus deformity range was from 1–38 degrees with a mean of 11.75 degrees. Haemoglobin (Hb) drop was calculated as a measure of blood loss. This was calculated using:

Hb drop = Pre-op Hb – (Day 3 Hb + No. of units of blood received)

(We assumed that one unit of blood equates to one gram increase in circulating haemoglobin).

Results: Osteoarthritis was the major diagnosis (90.3%) followed by rheumatoid disease (5.8%). Forty-four patients (21%) required no soft tissue release, this group having an average deformity of 6 degrees and an average Hb drop of 4.35g/dl. One hundred and forty-six (69%) patients required a posterolateral capsular release, average deformity of 14 degrees with an average Hb drop of 4.63g/dl. Seventeen (8%) patients had a posterolateral and an additional Iliotibial band release, average magnitude of deformity 18 degrees and average Hb drop of 4.97g/dl. One patient had a lateral collateral ligament release, magnitude of deformity 30 degrees and haemoglobin drop of 6.4g/dl.

Conclusion: Soft tissue release in total knee arthroplasties in the valgus knee increases the amount of postoperative bleeding and average length of stay of the patient.

Purpose: We describe our management of the valgus knee including release of tight lateral structures using a medial approach.

Method: Controversy surrounds which approach to use when undertaking a total knee replacement (TKR) in a fixed valgus knee. Poor patellar tracking is associated with this deformity and often rectified by a lateral retinacular release. Those favoring the lateral approach feel, although more demanding, it gives direct access to the tight lateral structures and avoids excessive devas-cularization to the patella, which may be associated with a lateral release and a medial arthrotomy.

Two hundred and eight consented patients (24.2%) were identified from 860 primary knee arthroplasties carried out over a 33-month period by a single surgeon. An LCS mobile bearing prosthesis was used in all cases. The mean valgus deformity measured 12.1 degrees (range 1–38). Fifty four percent of patients had a deformity of greater than 10 degrees. The patella was not resurfaced in any patient.

Results: Forty-four patients (21%) required no soft tissue release. The mean deformity was 6 degrees (range 2–13). Of the remaining 164 patients, 142 (87%) had a posterolateral capsule release, 17 (10%) posterolateral capsule and iliotibial band release, 4 (2.4%) posterolateral and direct posterior capsule release and 1 (0.6%) a lateral collateral ligament slide for fixed valgus deformity. The mean valgus deformity increased with each additional release required.

Of note 61 (29.3%) patients required a lateral patellar release for patellar maltracking. No patellar complications were noted. Mean patellar tilt was 1.1 degrees (sd=0.6 degrees) and mean patellar congruency 98% (sd=0.7%).

Conclusion: Using a medial approach in the valgus knee is technically less demanding than a lateral approach, can be used in any primary knee irrespective of the type of deformity and can restore patellar alignment without compromising viability at least in cases where the patella is not resurfaced.

Recent studies, including the Swedish Hip Register, have confirmed that modern cementing techniques are important to achieve long-term implant survival. Their ultimate goal is to obtain an increased strength of cement and its interfaces with bone, thereby maintaining secure fixation and effective load transmission.

The objective of this study was to measure the medullary pressures generated during bone cement injection, pressurisation and femoral prosthesis insertion for total hip arthroplasty. The measurements were recorded throughout the length of an in vitro femoral model while implanting a series of prosthetic hip stems using different pressurisation techniques. The prostheses used were the Charnley 40 flanged stem, an Exeter No. 3 stem, and a custom primary femoral component used in Belfast (Johnson & Johnson, DePuy International Ltd.). The following parameters were derived from the pressure data recorded; peak pressure, decay pressure and duration above 76 kPa, the pressure regarded as the threshold to obtain adequate bone penetration.

The range of peak distal stem pressures expected for all stems was 200–500kPa. The custom and Exeter stems generated proximal cement pressures in the range 100–300kPa. These pressures were attained through cement containment by stem design or auxiliary pressurising devices, respectively. It was observed that the Charnley femoral component did not perform as well with regard to proximal pressurisation, irrespective of which pressurisation technique was implemented. The durations of pressure maintenance above 76kPa are also important, 5 seconds being accepted as a minimum for an effective interface. These results reflect the pressure measurements, with adequate durations being maintained by those stems and pressurisers that were characterised by higher peak pressures.

It is concluded that stem design and the complementary cement management techniques are essential to realise the pressure/time characteristics that are regarded as necessary to form an optimum bone/cement interface.

Total knee arthroplasty has evolved considerably over the last thirty years. Early implant design achieved the short-term goals of pain relief and mobility, however loosening and polyethylene wear associated with over constraint was problematic. The Low Contact Stress total knee arthroplasty was developed in an attempt to address the problems of loosening and polyethylene wear. The highly congruent interface between the femoral component and the mobile insert minimises stress within the polyethylene and reduces the potential of wear and damage. Furthermore, the mobile bearing phenomenon minimises both torsional and shear stresses at the component bone interface. In our unit the impact of choice is the LCS rotating platform prosthesis, which is inserted with cruciate-sacrifice.

We reviewed 219 patients (272 knees) with an average follow-up of 6 years (5–8 years). In almost all cases the components were inserted with cement fixation. The patella was primarily resurfaced in 20 patients (21 knees). All operations were performed or supervised by the senior author. Female to male ratio was 2:1. Average age at surgery was 68 years (40–86) with osteoarthritis being the commonest primary diagnosis (89%). Postoperative range of motion ranges from 30–130° (average 103°). Average Oxford Knee, American Knee Society Score and Patellar Score was 19 (12–53), 160 (42–199) and 25 (4–30) respectively. Six patients (1.7%) required MUA at six weeks. Two patients (0.6%) required secondary patellar resurfacing. Three patients (0.8%) had revision of their components for persistent pain. At operation all components were noted to be well fixed. Spinout of the rotating platform occurred in one patient (0.3%). This was treated by exchange of the insert.

In conclusion, our early results of the LCS rotating platform prosthesis are encouraging with no cases of component loosening to date. This supports the continued use of the implant.

Currently, all details regarding implants are entered into a real time application on the Musgrave Park Hospital site using the Belfast Orthopaedic Information System (BOIS). This is a visual basic client application with data being stored in an SQL server database. This data collection system operates throughout every location within the hospital including the theatre block.

Loss of continuity occurs however when joint replacement takes place in Musgrave Park Hospital and then revision surgery or other procedures are carried out at other locations. The goal therefore of the Northern Ire-land Implant Register is to collect information on all implants performed and their revisions regardless of their location.

The dataset collected is based on the work of the National Joint Replacement Registry. Our system is designed to support and extend that dataset to provide a more comprehensive joint replacement registry database. This means that reports can be provided to individual sites and data entered into the National registry if required at a later stage.

At present every hospital in Northern Ireland is connected to the HPSS network (similar to the HPSSNet in the UK). What has been developed is a web browser based front end, which requires no complex software installation on any client machine. From this web based tool, staff at other locations can access information held at Musgrave Park Hospital, they can select an implant, or record a new implant and then link any revisions or other procedures carried out.

As all of the information is entered directly into the BOIS database, there is no delay in the information being available to all who access the system. This reduces the need for case notes to be transferred to other sites and the need to contact the other site directly regarding the case. The interface also provides a comprehensive reporting capability so that commonly requested standard reports are available for authorized staff to run from their web browser.

We present an overview of how the web interface works in practice and how data is entered into the system.

To maximise the long-term survivorship of any hip prosthesis it is important to recreate joint centre. Normal joint centre is determined by horizontal offset and vertical height of the acetabular and femoral components. In this study joint centre and horizontal offset were analysed in 200 consecutive patients operated on from October 1998 in whom the opposite hip was normal. Joint centre was defined relative to the acetabulum and femur both pre- and post-operatively. On the acetabular side a horizontal line was drawn across the pelvis immediately below each teardrop. A vertical line was drawn at right angles through the middle of each teardrop. Acetabular offset was defined as the horizontal distance from the vertical trans teardrop line to head centre. For femoral offset a screened x-ray was taken to show maximum offset. The anatomical axis was drawn and the offset was defined as the distance from the anatomical axis to head centre.

Our results show on the acetabular side there was an overall tendency to leave the joint centre medial and so decrease acetabular offset. However, we found that 90% of our sockets were placed within 6 mm of normal joint centre. We attribute this accuracy to the principle of visualising the transverse acetabular ligament intra-operatively and using this landmark to control depth of socket insertion. Conversely, on the femoral side there was a slight tendency to increase the offset. Nevertheless, 98% of the custom stems were within 10mm of normal joint centre. When we looked at total horizontal offset i.e. the combination of femoral and acetabular offset we found that joint centre had been restored to within 10mm in 93% of cases.

This study confirms the effectiveness of the custom femoral stem and Duraloc socket in restoring joint centre.

Rapidly progressive cases of primary idiopathic hip osteoarthrosis are well known and recognised. The prevalence reported in the literature varies from 4–18%. Three types have been identified- type 1 (rapid), type 2 (moderate) and type 3 (delayed) depending on the duration of chondrolysis and the subsequent rate of bone loss per year.

We reviewed the charts of all patients deemed to be RPO type 1 who had underwent hip arthroplasty under the care of the senior author (DEB) over a two-year period in an attempt to identify risk factors, which may have contributed to the rapid progression of their disease. All patients were treated using a custom femoral stem and a spiked Duraloc cementless socket following careful preparation of the acetabulum.

We identified 34 patients (40 hips) with type 1 rapidly progressive osteoarthrosis. Over the same time period 991 patients had underwent primary total hip arthroplasty, giving a prevalence of 4%. Of the 34 patients, 29 were female of average age 70.6 years (range, 51–83 years). All of the bilateral cases (6 patients) were female. Body mass index (BMI) for the female group ranged from 20.6 to 41.1Kg/m² (average, 28.2kg/m²) whilst that for the males was on average 25.8Kg/m² (range, 23.4–29.7Kg/m²).

Preoperative erythrocyte sedimentation rate (ESR) was 18mm/hr on average for the female group (range, 2–65mm/hr) and ranged from 3–52mm/hr (average, 20mm/hr) for the male patients. The preoperative Oxford Hip Score averaged 51 points for the female group and 48 points for the male group.

A detailed review of occupational history did not reveal any common occupational hazard. The majority of patients were non-smokers and denied any regular alcohol intake. Twenty-two patients (65%) had a history of hypertension. Twenty-seven patients (79%) had a history of non-steroidal anti-inflammatory use (most common preparation-diclofenac). Twenty-four patients (71%) resided in a rural area.

When compared to a cohort of patients undergoing primary total hip arthroplasty over the same time period, the only statistically significant risk factor identified was female gender.

We conclude, that patients who develop rapidly progressive osteoarthrosis of the hip are difficult to identify due to the absence of specific clinical features. We also outline our experience in the management of these technically challenging cases.

Many factors have been demonstrated to influence the range of knee movement that an individual can achieve.

The purpose of this study was to objectively demonstrate how range of knee movement is affected when the influence of pain is abolished.

Sixty-eight patients with degenerative joint disease presenting for primary total knee arthroplasty were recruited. Using a digital camera, images were taken before and after the induction of anaesthesia with the lower limb in four positions- extension, forced extension, flexion and forced flexion. Camera set up was standard and the range of knee motion was measured from the digital images.

Average arc of motion before anaesthesia was 96° (range, 41°–157°). After induction of anaesthesia, the arc of motion increased to 115° (range, 410–161°). Knee extension improved by an average of 5° (range, 0–15°) and flexion improved by an average of 16° (range, 0–65°).

In conclusion, these results demonstrate that pain has a significant inhibitory effect on the measured range of knee movement before surgical intervention.

Introduction: Since 1992 over 3000 custom-made cemented titanium femoral components have been implanted during total hip replacement in our centre.

Stems are machined using CAD-CAM. Measurements are made from screened AP and lateral x-rays of known magnification. Normal joint centre is recreated by controlling offset and vertical height of the femoral component.

Method: Joint centre and limb length were analysed radiologically in consecutive 100 patients following total hip replacement for unilateral arthritis. Joint centre was defined relative to the pelvis and femur.

The anatomical axis and offset of the femur were defined using a screened x-ray of known magnification taken to show maximum offset. Femoral centre height was defined relative to the greater trochanter.

Results: In general, acetabular joint centre was placed medial and high, tending to reduce limb length slightly. Conversely, on the femoral side the tendency was to leave the component proud, producing an increase in limb length. Most patients had limb length restored to within 6mm of normal.

This study confirms the effectiveness of the Belfast Custom Stem in restoring joint centre and limb length.

Dislocation of the polyethylene-rotating platform is a recognised complication of LCS knee arthroplasty.

We report ten cases of rotating platform dislocation out of 2151 primary total knee arthroplasties (0.5%) performed to date in our unit.

Of the ten cases (3 male: 7 female), six patients had a preoperative valgus deformity, two had a varus deformity and the remaining two patients were in neutral alignment, although the wear was predominantly within the lateral compartment. Two patients also had a previous patellectomy on the side of the platform spinout.

Of the ten cases, six patients were symptomatic as a result of their platform dislocation. The remaining four patients were asymptomatic and mobilising without any difficulty. In these patients, the dislocated platform was diagnosed on x-ray at outpatient review. Time to detection of the platform dislocation ranged from six days to two years.

Three patients required revision of their original insert to a larger, deeper dish insert. Three were managed by open reduction of the original insert. Of the remaining patients, one was managed successfully by closed reduction, one required an arthrodesis and one had the tibial insert cemented to the tibial tray. All patients at latest review have a functional and stable knee joint.

In conclusion, we feel that surgical error was to blame for the majority of our ten cases. Furthermore, we emphasise the importance of producing equal and balanced flexion and extension gaps at the time of knee arthroplasty in order to prevent mobile bearing dislocation. A novel technique for reducing a dislocated rotating platform is also described.

Periprosthetic fractures are a recognised complication of total knee arthroplasty. They may occur intraoperatively or postoperatively and risk factors have been identified which may predispose an individual to such a complication.

We report seven cases of periprosthetic tibial fractures following LCS total knee arthroplasty, a complication only encountered by the senior author following a change in practice from a cemented implant to a cementless one.

In light of this previously unreported complication in our unit, we have attempted to identify common features within this group of patients, that may have contributed to fracture occurrence.

Statistical analysis revealed a highly significant risk of periprosthetic tibial fracture in those patients with a preoperative neutral or valgus knee. Age, gender and diagnosis did not appear to significantly increase the risk of fracture. All patients displayed evidence of reduced bone mineral density in the lumbar spine and femoral neck regions on dual energy x-ray absorptiometry scanning.

We feel that those patients with a preoperative neutral or valgus knee and local evidence of osteopenia represent a high-risk group, in whom particular care in alignment should be taken. In this group it may also be preferable to have the tibial component inserted with cement fixation.

Isolated patello-femoral osteoarthritis has been reported to occur in approximately 5% of patients with gonarthrosis of the knee. We present the results of the surgical management of this disorder in our unit. Thirty-one patients (33 knees) with symptomatic patello-femoral osteoarthritis and minimal tibio-femoral changes underwent LCS total knee replacement without patellar resurfacing.

Average age was 73 years (range, 58–89 years) with a female to male ratio of 5: 1. Average follow-up was 20 months (range, 12–40 months). All except four patients had grade 3 or 4 patello-femoral osteoarthritis according to the classification by Sperner et al.

Preoperatively, all patients had significant knee pain. Sleep disturbance was reported in 21 patients. All but 10 patients required walking aids. Average range of motion was 108° (range, 80°–125°).

At latest review, 21 knees are pain-free, the remaining 12 knees describing only occasional knee pain. Two patients continue to have night pain. Average range of motion was 104° (range, 700 – 135°). Lateral patellar tilt improved in all but five knees by an average of 7° (range, 10–26°). Patellar congruency improved in all but three knees by an average of 18% (range, 3–63%). None of the patients to date have required revision surgery. We suggest that knee arthroplasty without patellar resurfacing is an effective option in older patients with isolated patello-femoral osteoarthritis.