Knee Surg, Sports Traumatol, Arthroscopy (1993) 1 : 68-70 Knee Surgery Sports Traumatology Arthroscopy

�9 Springer-Verlag 1993

Originals

The use of continuous passive motion after arthroscopically assisted anterior cruciate ligament reconstruction: help or hindrance?

G.E.Witherow, S. R. Bollen, L. A. Pinczewski

Australian Institute of Musculoskeletal Research, Sydney, Australia

Abstract. One hundred and eight patients having under- gone arthroscopic anterior cruciate replacement and having had continuous passive motion (CPM) as part of their immediate post-operat ive regime were prospective- ly compared with 108 patients having an identical opera- tive procedure but not receiving CPM. All patients were operated on by the same surgeon, and the two groups were well matched for age, weight and associated injuries and procedures. Those not receiving CPM required sig- nificantly less analgesia (P = 0.0001), had less blood loss measured in the drains (P = 0.001) and had a shorter hospital stay (P = 0.0001). At review 6 months after sur- gery, there was no significant difference between the two groups in the range of movement of the operated leg compared to the normal leg.

Key words: Arthroscopic anterior cruciate ligament re- placement - Continuous passive motion (CPM)

Introduction

Continuous passive motion (CPM) as part of the im- mediate post-operat ive care is now part of the recom- mended regime of several operative methods of anterior cruciate l igament (ACL) reconstruction [4, 5]. It is said to have benefits when compared to traditional techniques of immobilisation or delayed mobilisation, in that it reduces quadriceps wasting without increasing analgesic requirements, joint effusion, haemarthrosis or soft tissue swelling, after both open and arthroscopic procedures [6].

Recent debate has been about the length of t ime that CPM needs to be used for post-operatively. It has been shown that there is no measurable difference in a variety of parameters assessed at 42 days after surgery whether CPM was used for 4 days or 14 [8].

Correspondence to: S.R.Bollen, MB FRCS FRCSEd Orth, c/o Department of Orthopaedics, Bradford Royal Infirmary, Duck- worth Lane, Bradford BD9 6R J, UK

One of the present authors (L.A.P.) noted an appar- ent increase in patients ' post-operat ive pain and blood loss when CPM was used in the post-operat ive regime, at odds with previous reports [4, 6]. A study was set up to investigate whether using CPM in the immediate post- operative period influenced analgesic requirements, post-operat ive blood loss and length of hospital stay. Range of movement in the operated leg compared to the normal leg was measured at 6-month review, at which time return to sporting activities was allowed.

Patients and methods

Over the time period June 1990 to April 1991, 108 consecutive patients undergoing arthroscopic ACL reconstruction and having CPM in the immediate post-operative period were prospectively compared with 108 patients undergoing an identical operative pro- cedure but without CPM in the immediate post-operative period. Patients were referred to the clinic from all over Australia with mainly sports-related injuries, most commonly from rugby and ski- ing. At the height of the season, up to 18 ACL reconstructions a week were being performed. The only patients excluded from the study were those having revision reconstructions or combined anterior and posterior cruciate reconstruction.

There were 20 acute cases (less than 3 weeks post-injury) and 88 chronic injuries (more than 3 weeks post-injury) in the group receiving CPM and 23 acute injuries and 85 chronic injuries in the non-CPM group.

All operations were performed by one surgeon (L.A.P.). Re- construction ufilised the middle third of the patella tendon with interference fit screw fixation, in a slightly modified version of the method first described by Rosenberg et al. [7]. "White zone" bucket handle and flap tears of the menisci were excised and peripheral tears were repaired with an inside-out technique using Maxon sutures. All patients had one intra-articular and one extra- articular drain. After induction of general anaesthesia all patients received a femoral nerve block with 20ml 0.5% bupivacaine with adrenaline. After surgery all patients received four doses of naproxen 500 mg orally.

The post-operative regime for the group receiving CPM was 24 h CPM (the model used was manufactured by the Toronto Med- ical Corporation), started in the recovery room. The calf and ankle were supported and the range of movement set from 0 to 60 ~ Inner range quadriceps and hamstrings exercises and co-contrac-

69

tions were performed in concert with the machine, supervised by a physiotherapist. On the 2nd post-operative day, CPM was discon- tinued and active exercises begun as in the non-CPM group.

In the non-CPM group, inner range quadriceps and hamstrings exercises were started on the 1st post-operative day and active flexion allowed, aiming to achieve 90 ~ . Twenty-four hours post- operatively the patients were mobilised, bearing weight as pain al- lowed and using crutches if required, In both groups the drains were removed at 24h. Only patients with an associated grade 2/3 medial collateral ligament injury were issued with a brace.

In each group the number of doses of post-operative injectable and oral analgesics giveJ: was recorded. Dosage was taiIored to the size of the patient and was decided by one of a group of three anaesthetists who anaesthetised all the patients and were unaware of the study. The proportion of each group anaesthetised by each anaesthetist was very similar. Blood loss collected in drainage bottles was measured, and the length of post-operative hospital stay was recorded. Patients were reviewed 6 months after opera- tion, and any loss of extension or flexion in the operated leg com- pared to the normal leg was recorded.

Statistical analysis of the results was carried out by the Depart- ment of Statistics at Sydney University, using the unpooled two- sample t method.

Results

The mean age in the non-CPM group was 28.6 years and the mean weight was 74.75 kg. There were 66 males and 42 females with 60 left and 48 right injured knees. The mean age in the CPM group was 28.3 years and the mean weight 73.4 kg. There were 74 males and 34 females with 53 left and 55 right injured knees.

The overall numbers of associated meniscal injury requiring excision or suture and of associated chondral or medial collateral l igament injury were very similar in the two groups (Table 1).

The results and statistical significances are sum- marised in Table 2. It can be seen that there was a signif- icantly lower mean injectable and oral analgesic require- ment (1.34 vs 2.34 doses and 2.6 vs 3.6 doses respective- ly) in the group not having CPM. There was also signifi- cantly less mean measured blood loss (187.1 vs 233.7 ml) and a shorter hospital stay (2.42 vs 2.94 days) in this group.

Twenty percent of each group failed to attend for review at 6 months. Of the 80% of each group attending, seven patients in each group failed to achieve full ex- tension of the operated leg compared to the normal leg

Table 1. Distribution of associated injuries and surgical procedures

CPM group Non-CPM group

Acute Chronic Acute Chronic n n n n

Medial collateral injury 3 Medial meniscectomy 3 Lateral meniscectomy 7 Medial meniscal suture 2 LateraI meniscaI suture 2 Significant chondral injury 0

0 0 1 25 2 26 17 6 23 3 6 6 4 1 9 2 2 7

CPM, Continuous passive motion

Table 2. Comparison of analgesia requirement, blood loss and hos- pital stay in CPM and non-CPM groups

CPM Non- P group CPM

group

Mean number of doses intramuscular analgesia 2.3 1.3 <0.0001

Mean number of doses oral analgesia 3.6 2.6 <0.0008

Mean length of hospita! stay (days) 2.9 2.4 <0.0001

Mean blood loss (ml) 233,7 187.1 <0.001

Statistical analysis by the unpooled two-sample t method

(including any physiological recurvatum). No patient in either group had lost more than the last 5 ~ of extension.

Eleven patients in the CPM group (seven of whom were also the patients who failed to regain full extension) and seven patients in the non-CPM group (three of whom also failed to regain full extension) failed to regain ful! flexion. In the CPM group sever~ patients tost 5 ~ of flexion, two 5-10 ~ and two 10-20 ~ In the n o m C P M group, five patients lost 5 ~ and two patients 10-20 ~ of flexion. The differences were not statistically significant.

There were three re-ruptures diagnosed arthroscopi- cally in the CPM group occurring within 6 months of operation.

Discussion

Continuous passive motion has an established place following total knee replacement [3] and articular carti- lage injury [9] but following A C L reconstruction the evi- dence is less clear. Although Noyes et al. [6] apparently showed clear advantages in terms of preventing quadri- cops wasting with no increase in analgesic requirements, hospital stay, soft tissue swelling or knee effusion, the numbers in each group of their study were so small (4 or 5 in each group) that it is difficult to draw any valid sta- tistical conclusions.

Surgery for A C L reconstruction can now be per- formed arthroscopically, minimising joint t rauma. Inter- ference screw fixation, providing stable anchorage for grafts, when coupled with accurate, isometric graft placement, enables early active motion through a full range of movement without causing wound problems or early graft loosening (as repor ted by Burks e t a l . [2]), and minimises any long-term problems with loss of nor- mal joint motion. The decrease in pain following ar- throscopic surgery compared to open reconstruction de- creases quadriceps inhibition [1], which, together with early active motion, helps minimise muscle wasting, allowing a more rapid return to normal function.

As operat ive techniques have improved, allowing a more rapid recovery, and pressures on bed occupancy continue to increase, so length of hospital stay has inevit- ably decreased. This has meant that the time available for post-operat ive use of CPM has become increasingly

70

limited. This study addressed the quest ion of whether short periods of pos t -opera t ive CPM are advantageous to the patient.

The results of this study contradict the belief that CPM is of any benefi t in the immedia te post -operat ive period, or has any influence on the results at 6-month follow-up, after an ar throscopic A C L reconstruct ion of this type. Indeed , it increases analgesic requirements , measured blood loss and length of hospital stay.

On the basis of this s tudy we believe that there is no longer a place for the rout ine use of CPM in the im- mediate post -operat ive period following ar throscopic A C L reconstruct ion with an active post -operat ive mobil isat ion regime.

Acknowledgements. We would like to thank Dr John Edelman of the Department of Statistics, Sydney University, for his help in the statistical analysis of our data.

References

2. Burks R, Daniel D, Losse G (1984) The effect of continuous passive motion on anterior cruciate ligament reconstruction sta- bility. Am J Sports Med 12 : 323-327

3. Ecker ML, Lotke PA (1989) Postoperative care of the total knee patient. Orthop Clin North Am 20 : 55-62

4. Hamilton HW, Hoffman DV, Morris JS, Porter JS (1988) Con- tinuous passive motion in postoperative knee rehabilitation. In: Feagin JA Jr (ed) The crucial ligaments. Churchill Livingstone, Edinburgh, pp 465-471

5. Leeds Keio System 90. Instructional literature. Howmedica 6. Noyes FR, Mangine RE, Barber S (1987) Early knee motion

after open and arthroscopic anterior cruciate ligament recon- struction. Am J Sports Med 15 : 149-160

7. Paulos LE, Cherf J, Rosenberg TD, Beck CL (1991) Anterior cruciate reconstruction with autografts. Clin Sports Med 10: 469-485

8. Richmond JC, Gladstone J, MacGillivray J (1991) Continuous passive motion after arthroscopically assisted anterior recon- struction: comparison of short versus long term use. Arthros- copy 7 : 39-44

9. Salter RS, Simmonds DE, Malcolm BW, Rumble EJ, Mac- Michael D, Clements NMD (1980) The biological effect of con- tinuous passive motion on the healing of full thickness defects in articular artilage. J Bone Joint Surg [Am] 62 : 1232-1251

1. Arvidsson I, Eriksson E, Knutsson E, Arner S (1986) Reduc- tion of pain inhibition on voluntary muscle activation by epi- dural analgesia. Orthopaedics 9:1415-1419