Introduction

Surgical treatment of symptomatic bilateral knee arthritis has led to much debate amongst orthopaedic surgeons and the literature appears to be split for (₁,₂,₃,₄,₅,₆,₇,₈,₉,₁₀,₁₁,₁₂,₁₃,₁₄,₁₅,₁₆,₁₇,₁₈,₁₉,₂₀,₂₁) or against (₂₂,₂₃,₂₄,₂₅,₂₆ ).

The advantages of a simultaneous bilateral knee arthroplasty include a single anaesthetic, reduced hospital stay and cost effectiveness. These are counterbalanced by a perceived greater risk due to a longer anaesthetic, pain relief, blood requirements, and thrombus or fat emboli.

In this retrospective study, two similar groups of patients were compared according to the result obtained after undergoing a unilateral or bilateral knee arthroplasty by the same orthopaedic firm. Length of hospital stay, blood requirements, and change in haemoglobin values, analgesia requirements and complications were recorded and compared.

Subjects and Methods

Data was collected from patients that underwent from 1999-2002 underwent unilateral or bilateral total knee replacements by the same orthopaedic team at St Luke's General Hospital in Malta. Of these 20 underwent bilateral total knee arthroplasties whilst 21 underwent unilateral procedures.

Patients underwent a routine pre-operative medical assessment and where warranted were assessed by a physician and anaesthetist and counselled as to the risks involved. In this study there was no upper age limit for a unilateral or bilateral procedure. The oldest patient in the unilateral group was 78 years and in the bilateral group 83 years.

Regional (spinal) anaesthesia was used in 7 of the bilateral group and 13 of the unilateral group, while general anaesthesia was used in the remainder. Tourniquet was used in both groups and a single preoperative dose of cefuroxime 1.5gram given intravenously, followed by 2 further doses of 0.75 grams 8 hours and 16 hours post operatively. A standard medial parapatellar approach was employed. Excision of osteophytes and balancing of soft tissues and adequate realignment to correct any angular deformities were performed. Similar design posterior cruciate-sparing TKR prosthesis was utilised for both groups and cemented using third generation techniques. Two suction drains were inserted in each knee. Skin closure was followed by sterile dressing and compression bandaging around the knee.

Both groups of patients were treated according to a similar post operative protocol of compression stockings and low dose unfractionated heparin. Physiotherapy included static quadriceps excercises on day 1 and assisted knee flexion. Suction drains were removed on day 2 post op.

From the data collected in the medical files, hospital stay, blood usage, perioperative haemoglobin values, time to ambulate, complications and analgesia were monitored.

Results

The following is a table showing patient demographics:

Figure 1

Table 1: Distribution of groups

An unexpected finding was that when the results were grouped together, males were statistically significantly (p<0.05) younger 68.9 years (SD4.9) as a group when compared to females 72.6 years (SD4.5).

Length of stay

Blood usage

Blood transfusions were required in 29% (n=6) of the unilateral group and 50%(n=10) of the bilateral group. However this was not statistically significant (p=0.16).Neither was there any statistically difference between the pre- and post-op haemoglobin values between the groups (p=0.35).

Figure 2

Table 2: Variation in pre- and post-op haemoglobin values

Days to ambulate

Ambulation was considered as the patient walking with the aid of a therapist. There were no statistical differences between the groups (p=0.46).

Figure 3

Table 3: Days to ambulate

Analgesic requirements

Various analgesia preparations were utilised post-op including opiods (pethidine), NSAIDS, paracetamol and co-proxamol. Typical post-op analgesia regime included patient controlled analgesia with intra-venous opioids for the 1st one to two days post op, followed by oral analgesics with intramuscular opioids used only for break-through pain. However requirements were tailored according to patient demand. Regarding length of analgesic use post-op there were no statistical significant variations between the groups (p=0.78)

Figure 4

Table 4: Length of analgesia use

Complications

Excluding blood transfusion, a number of post-op complications occurred in the unilateral (n=5) and bilateral (n=5) groups. These ranged from DVT to an episode of TIA.

Figure 5

Table 5: Post-op complications

Discussion

From this study simultaneous bilateral knee arthroplasty appeared to be safe and the alleged risks involved were more than acceptable. This finding compares well with a number of other studies [₁,₂,₃,₄,₅,₆,₇,₁₀,₁₁,₁₂,₁₃,₁₄,₁₅,₁₆,₁₇,₁₈,₁₉,₂₀,₂₁].

The average length of stay for the bilateral procedure patients was 1.7 days longer. These indicate an important reduction in bed occupancy and therefore accompanying costs and these results are seen in other studies [₁,₂,₄,₈,₉,₁₁,₁₂,₁₅,₂₁].

Blood usage and peri-operative change in haemoglobin indices indicate a trend for increased blood loss in the bilateral procedure. This would reflect the findings in various studies some of which even obtained statistically significant results [₁,₁₂,₁₅,₂₂] whilst others showed no differences between the groups [₂,₃,₄,₅,₆,₇,₁₁,₁₃,₁₄,₁₅,₁₆,₁₇,₁₈,₁₉,₂₀,₂₁].

It therefore appears that patients for total knee arthroplasties, especially if bilateral, are ideal candidates for autologous blood autotransfusion use during the operation and for autologous pre-donated blood.

Days to ambulate were practically identical in both groups. This finding was documented in other studies [₁,₂,₄,₆,]. This is an important finding as early mobilisation of the patient is fundamental in the rehabilitative process and to minimise further the risk of thrombo-embolic complications.

As in other studies [₁,₆,₁₂,₁₇,₁₈] length of analgesic use was practical the same between the two groups. A recent study has advocated the use of an epidural in the bilateral group [₁] but this was not supported by other literature.

Complication rates were comparable between the groups and this is supported by a large number of studies [₁,₂,₃,₄,₅,₆,₇,₁₀,₁₁,₁₂,₁₃,₁₄,₁₇,₁₈,₁₉,₂₀,₂₁]. However, other studies have indicated an increase in postoperative complication in the bilateral group, especially cardiopulmonary and post-operative confusion [₁₅,₁₉,₂₂,₂₃,₂₄,₂₅,₂₆].

In conclusion, the decision to perform a bilateral versus unilateral knee arthroplasties must take into account three factors: the patient's desires; the medical condition of the patient when one considers the longer anaesthetic and twice the physical trauma; and lastly in this modern day the availability of beds and the cost saving advantages of a bilateral procedure [₄,₈,₉,₁₄,₁₅,₂₀,₂₃].

Correspondence to

Name Mr. Ryan Giordmaina Address Department of Orthopaedics, Mater Dei Hospital, B'Kara, Malta, Europe Phone numbers +356 99293201, +356 2545 5226 E-mail address rgiordmaina@yahoo.co.uk