Printer Friendly

Continuous intra-articular infusion of ropivacaine after unilateral total knee arthroplasty.

There are a variety of techniques used in the management of postoperative pain after total knee arthroplasty. Single-shot regional blocks provide good pain control in the immediate postoperative period but rarely persist beyond 24 hours. The efficacy of continuous femoral nerve infusion over opioid regimens is well-established and approximates that of epidural infusion for postoperative analgesia (1,2). It has advantages over the epidural route with a reduced level of 'complexity' of care, complications and easier mobilisation, especially of the non-operative side.

Continuous wound instillation with local anaesthetic has enjoyed some popularity for many procedures with the advent of reliable elastomeric pumps and multi-hole catheters. A potential advantage of intra-articular infusion of local anaesthetic is that it allows early physiotherapy without motor blockade of the quadriceps muscle. However, its efficacy is yet to be established. A retrospective study involving intra-articular infusion of local anaesthetic after knee arthroplasty yielded encouraging results with reduced opioid consumption but did not involve any regional anaesthetic technique (3).

The aim of this study was to compare the efficacy of intra-articular local anaesthetic with placebo infused over 48 hours after unilateral total knee arthroplasty. We were particularly interested in the time period from 24 to 48 hours postoperatively as, in our experience, this is the time period of maximum discomfort when the intraoperative nerve blocks have worn off. A secondary aim was to look for evidence of a dose effect by using two different concentrations of local anaesthetic.


After approval from the Tasmanian Human Research Ethics Network, 66 consenting adults were enrolled into the study. Patients were allocated to one of three groups using permuted block randomisation with blocks of eight (ratio 2:1:1 controls : low-dose : high-dose) with a final block of four. The randomisation occurred after surgery had commenced and the process took place in the pharmacy department, where the 270 ml elastomeric balloon of an On-Q[R] PainBuster[R] 5 ml/hour 6.5 cm soaker catheter delivery system (I-Flow Corporation, California, USA) was filled with 240 ml of normal saline (control), 0.2% ropivacaine (low-dose) or 0.375% ropivacaine (high-dose). Group allocation was known only by pharmacy staff who took no other part in the trial. Allocation and group identity were only revealed after all data had been collected and checked for consistency. Where allocations were made and no data collected, the allocation was reissued by the pharmacy at a later date to ensure that the numbers in each group were balanced.

A total of 70 patients were assessed for eligibility. Two were excluded (one having bilateral knee replacements and the other with a marked intolerance to morphine) and two refused. Thus, 66 patients were randomised. The final six received the allocations of patients who were randomised but on whom no postoperative data were collected due to administrative reasons. Data assumed to be missing for one patient (in the high-dose group) were actually misplaced, so there was an additional allocation to that group.

Demographic data collected were age, gender, weight and ASA physical status score. All patients underwent general anaesthesia. The technique was not fixed but precluded the use of ketamine or high-dose opioids. Prior to induction, femoral and anterior sciatic nerve blocks were performed using a nerve stimulator. Ropivacaine 0.75% 100 mg was used for each block. The surgical approach was similar for all patients (medial parapatellar) and an intra-articular drain and the PainBuster[R] catheter were both sited into either side of the joint by the surgeon prior to closure of the knee capsule. The elastomeric balloon (pre-filled in pharmacy) was attached and the infusion commenced in the post-anaesthesia care unit (PACU). Data recorded included operation time, PACU time until ward readiness, intraoperative morphine administered and PACU morphine required. Ability to obtain nerve stimulation during performance and evidence of block success assessed in PACU (motor block, sensory block to ice) were recorded for each block on a three-point scale (certain, not certain, unable) and block success defined as all "certain" or a maximum of one "uncertain".

Patients were prescribed intermittent subcutaneous morphine pro re nata (prn) according to a protocol that allows a dose frequency as short as one hour. The use of oral opioid analgesia was encouraged and oxycodone was available two-hourly on demand. Virtually all patients received regular paracetamol, but non-steroidal analgesics were precluded in the study design as they are not uniformly tolerated and are contraindicated in some patients.

Patients were visited 24, 48 and 72 hours after surgery and the following data recorded: verbal numerical rating scores for pain at rest, on movement, maximal pain intensity and average pain over the previous 24 hours. Subjective assessment of analgesic efficacy was recorded on a five-point scale in answer to the question "How effective do you think the catheter has been at relieving your pain?" with 1=excellent, 2=good, 3=neutral, 4=poor and 5=very poor/made it worse. Morphine and oxycodone consumption were also recorded, along with the presence of any major adverse events. The PainBuster[R] catheter was removed at 48 hours along with the articular drain (unless there was persistent drainage). Patient notes were reviewed at discharge to determine length of stay and the presence of any serious adverse events.

Statistical analyses

The sample size was based on previous studies (1-4) in which the average visual analogue score for pain on movement on day two was 50 mm (standard deviation 30) for patients with morphine alone. The average equivalent morphine consumption was approximately 30 mg on day two (standard deviation 10). Accepting a threshold P value ([alpha]) of 0.05 and a power of 80%, 16 patients per group would be required to detect a 30% reduction in morphine use and 23 patients would be required to detect a 25 mm reduction in visual analogue scale. Neither measure is strictly normally distributed. We elected to use verbal numerical rating scores rather than the visual analogue scale. Since we planned to use non-parametric statistics and to look at any efficacy by dose-dependence, we decided to enrol at least 60 patients (30 controls and 15 each to the low- and high-dose groups).

The analysis was predetermined to be in two stages. In stage one, primary end-points of movement pain and opioid consumption at 48 hours were compared between the control and treatment groups using a Wilcoxon rank test. If significant at P <0.05, an adjustment was made for age using ordinal logistic regression. In stage two the three groups were compared to look for an increasing effect of ropivacaine dose/concentration as stronger evidence of treatment effect. Pain scores, subjective assessment of analgesic efficacy scores and opioid consumption were compared using a Kruskall-Wallis test and the opioid consumption also analysed with ordinal logistical regression adjusting for age.

Other tests for group homogeneity and exploratory analyses included Student's t-test for continuous variables and Fisher's exact test for categorical variables. All analyses were performed using Stata/IC 10.0 (Stata Corporation, Texas, USA).


In all, there were data for 61 allocations. There were missing data of primary end-points in one patient in the high-dose group. The groups were demographically similar (Table 1), although the average age in the control group was older and they were more likely to have the procedure in the public hospital. The groups were well-matched on most variables recorded in the first 24 hours (Table 2). This was to be expected, as all patients had similar treatment and the regional nerve blocks performed would be anticipated to be effective for most of that time.

The second data epoch (24 to 48 hours) should have reflected any differences in the group due to the intervention, as all blocks would have been expected to have worn off. These results are shown in Table 3. The difference in oxycodone consumption was not significant (P=0.148) when adjusted for age in an ordinal logistic regression model. However, the secondary analysis demonstrated statistically significant increases for both pain scores and opioid consumption for the high-dose group vs control, even when age was taken into account. These results are shown in Table 4.

The final data epoch (48 to 72 hours) would have been expected to have few differences between groups since the intra-articular catheter was removed at 48 hours. Again there was greater oxycodone use in the treatment group (particularly the high-dose group) but not when adjusted for age (ordinal logistic regression, P=0.13). Finally, the lengths of stay were similar in each Group (Table 5).


There was no positive benefit of intra-articular infusion of local anaesthetic after knee replacement demonstrated in this study. Group differences from 24 to 48 hours would have been expected to be due to intra-articular local anaesthetic or random effects. Surprisingly, the group that had the higher concentration of ropivacaine had higher pain scores and required more analgesia.

This lack of efficacy is supported by other studies in the literature for intra-articular bupivacaine infusion after knee replacement (5), anterior cruciate ligament reconstruction (6,8) and arthroscopic rotator cuff repair (9). The latter study also used two different doses of local anaesthetic and noted a non-significant trend towards higher pain scores and opioid consumption in the group that received the larger dose. Only one of these studies used any form of regional anaesthesia (7). Wound infiltration combined with intra-articular injection of local anaesthetics has been shown to be superior to lumbar epidural analgesia after total hip arthroplasty (10). This may be more to do with the wound infiltration than the intra-articular infusion (11).

The overall lack of efficacy might be explained by the presence of wound drains. A previous study has suggested that drug loss through drainage may exceed 25% (5).

The methodology of this study differs from many similar studies in that patient-controlled morphine analgesia was not used as the comparator for opioid consumption. Since the introduction of a flexible subcutaneous morphine protocol at our hospital several years ago, the use of patient-controlled analgesia has declined markedly. Also, most patients are using oral analgesics only from 24 hours postoperatively. The pain scores and opioid consumption in both groups were consistent with previously published data that were used to generate the sample size calculation; therefore we do not believe that this study was underpowered. The use of adjustments for age as a covariant with small sample sizes may appear as 'over-analysis'. However, despite the apparent similarity of ages between the groups, age is such a strong predictor of opioid requirement" that small differences can lead to erroneous conclusions. We would argue that age should always be adjusted for in the analysis of opioid use (rather like using cardiac index rather than cardiac output).

This study did not measure functional outcomes. This was a deliberate omission based on the lack of uniformity of physiotherapy practices between the two hospitals and between surgeons in the applications of knee bandages. It was also felt that unless there were significant differences in pain experiences, the chance of different functional outcomes was remote.

It is difficult to explain the apparent poorer pain control in the higher dose ropivacaine group. This result may well be due to chance alone but there are other possible explanations including some unmeasured confounder(s). The patients in the treatment group were not only younger but more likely to have the operation performed in the public hospital. This might mean that they had waited longer in pain for their operation and that their pain was not isolated to the operative site.

Duration of effectiveness of the nerve blocks was not assessed and it is possible that patients in the high-dose group had less effective blocks than in the other two groups. There was greater evidence of block success in the control group. Exploratory analysis confirmed that block 'failure' was associated with significantly higher average movement and maximum pain scores in the first 24 hours (using linear regression) but not with opioid consumption. There was no correlation between block failure and any measured outcome in the 24 to 48 hour period.

We did not measure plasma concentrations but no signs or symptoms that could be explained by systemic toxicity were reported during the trial and previous studies have reported safe levels of plasma ropivacaine after extensive infiltration followed by intra-articular infusion (13).

The safety of intra-articular infusion in terms of potential joint infection has not been fully established. The published literature includes 126 retrospective cases with no infections but five cases of prolonged wound drainage (3,4).

The complication rate in this study (two infected knees in 61 cases) is higher than the international literature but the sample size is too small to draw meaningful conclusions (14,15). It would be a concern if it were to be replicated over a larger number of knee replacements and might reflect the presence of an intra-articular catheter.

In conclusion, we could identify no positive benefit of intra-articular infusion of local anaesthetic after total knee arthroplasty either in our study or in the literature.


We would like to thank the orthopaedic surgeons involved who agreed on a consistent approach, especially Dr Scott Fletcher who performed the majority of the cases. We would also like to thank members of the Department of Anaesthesia who collected much of the data. Finally, we would like to thank Ms Suzette Seaton and the Pharmacy Department, who performed the allocations and prepared the infusions.

Accepted for publication on May 1, 2009.


(1.) Chelly JE, Greger J, Gebhard R, Coupe K, Clyburn TA, Buckle R et al. Continuous femoral blocks improve recovery and outcome of patients undergoing total knee arthroplasty. J Arthroplasty 2001; 16:436-445.

(2.) Singelyn FJ, Deyaert M, Joris D, Pendeville E, Gouverneur JM. Effects of intravenous patient-controlled analgesia with morphine, continuous epidural analgesia, and continuous three-in-one block on postoperative pain and knee rehabilitation after unilateral total knee arthroplasty. Anesth Analg 1998; 87:88-92.

(3.) Skinner HB, Shintani EY. Results of a multimodal analgesic trial involving patients with total hip or total knee arthroplasty. Am J Orthop 2004; 33:85-92; discussion 92.

(4.) DeWeese FT, Akbari Z, Carline E. Pain control after knee arthroplasty: intraarticular versus epidural anesthesia. Clin Orthop Relat Res 2001; 226-231.

(5.) Nechleba J, Rogers V, Cortina G, Cooney T. Continuous intra-articular infusion of bupivacaine for postoperative pain following total knee arthroplasty. J Knee Surg 2005; 18:197-202.

(6.) Parker RD, Streem K, Schmitz L, Martineau PA. Efficacy of continuous intra-articular bupivacaine infusion for postoperative analgesia after anterior cruciate ligament reconstruction: a double-blinded, placebo-controlled, prospective, and randomized study. Am J Sports Med 2007; 35:531-536.

(7.) Alford JW, Fadale PD. Evaluation of postoperative bupivacaine infusion for pain management after anterior cruciate ligament reconstruction. Arthroscopy 2003; 19:855-861.

(8.) Dauri M, Polzoni M, Fabbi E, Sidiropoulou T, Servetti S, Coniglione F et al. Comparison of epidural, continuous femoral block and intraarticular analgesia after anterior cruciate ligament reconstruction. Acta Anaesthesiol Scand 2002; 47:20-25.

(9.) Banerjee SS, Pulido P, Adelson WS, Fronek J, Hoenecke HR. The efficacy of continuous bupivacaine infiltration following arthroscopie rotator cuff repair. Arthroscopy 2008; 24:397-402.

(10.) Andersen KV, Pfeiffer-Jensen M, Haraldsted V, Soballe K. Reduced hospital stay and narcotic consumption, and improved mobilization with local and intraarticular infiltration after hip arthroplasty: a randomized clinical trial of an intraarticular technique versus epidural infusion in 80 patients. Acta Orthop 2007; 78:180-186.

(11.) Bianconi M, Ferraro L, Traina GC, Zanoli G, Antonelli T, Guberti A et al. Pharmacokinetics and efficacy of ropivacaine continuous wound instillation after joint replacement surgery. Br J Anaesth 2003; 91:830-835.

(12.) Macintyre PE, Jarvis DA. Age is the best predictor of postoperative morphine requirements. Pain 1996; 64:357-364.

(13.) Stringer BW, Singhania AK, Sudhakar JE, Brink RB. Serum and wound drain ropivacaine concentrations after wound infiltration in joint arthroplasty. J Arthroplasty 2007; 22:884-892.

(14.) Peersman G, Laskin R, Davis J, Peterson M. Infection in total knee replacement: a retrospective review of 6489 total knee replacements. Clin Orthop Relat Res 2001; 15-23.

(15.) Wilson MG, Kelley K, Thornhill TS. Infection as a complication of total knee-replacement arthroplasty. Risk factors and treatment in sixty-seven cases. J Bone Joint Surg Am 1990; 72:878-883.

M. REEVES *, M. W SKINNER [[dagger]]

Departments of Anaesthesia, North West Regional and North West Private Hospitals, Burnie, Tasmania, Australia

* M.B., B.S., F.A.N.Z.C.A., P.G.Dip.Biostat., Visiting Medical Officer.

[[dagger]] M.Sc., Dip.D.H.M., M.B., B.S., F.A.N.Z.C.A., Director.

Address for correspondence: Dr M. Reeves, Department of Anaesthesia, North West Regional Hospital, Burnie, Tas. 7320. Email:
Demographic data

 Control (n=30) Treatment (n=31)

Age 72(10) 67(9)
Gender (M/F) 12/18 13/18
Weight 84(14) 91 (17)
Private/public 20/10 14/17
ASA score (I/II/III) 3/21/5 2/22/7

* Values expressed are mean (SD). M=male, F=female, ASA
score=American Society of Anesthesiologists physical status

The first 24 hours

 Control (n=30) Treatment (n=31)

Operation time (min) 103 (31) 114 (39)
PACU time (min) 63 (31) 57 (24)
Block success * 24 (80%) 22 (71%)
Total morphine (mg) 15 (9-18) 11 (5-27)
Oxycodone (mg) 15 (5-30) 15 (10-30)
 Rest 3 (2-5) 1 (0-4)
 Movement 5.5 (3-8) 4 (2-7)
 Maximum 7 (3-8) 6 (2-7)
 Average 3 (2-5) 2 (1-5)
 SAOAE 2 (1-3) 2 (1-3)

Time expressed as mean (SD). Drug doses are median
(inter-quartile range). * See text for definition.
PACU=post-anaesthesia care unit, VNRS=verbal numerical
rating scale for pain expressed as median (interquartile
range, IQR), SAOAE=subjective assessment of
analgesic efficacy as median (IQR).

Postoperative data from 24 to 48 hours

 Control (n=30) Treatment (n=30) P value
Rest 2 (1-4) 4 (2-5) 0.18
Movement 6 (4-8) 6 (4-8) 0.96
Maximum 6 (4-8) 8 (5-10) 0.13
Average 4 (3-5) 4 (3-5) 0.96
SAOAE 2 (1-4) 3 (2-3) 0.42
Morphine (mg) 1.5 (0-12) 0 (0-15) 0.65
Oxycodone (mg) 20 (10-40) 30 (15-50) 0.04

VNRS=verbal numerical rating scale for pain expressed as median
(IQR), SAOAE=subjective assessment of analgesic efficacy as
median (IQR). P values are from Wilcoxon rank sum test.

Postoperative data from 24 to 48 hours. Secondary analysis

 Control LD HD P value
 (n=30) (n=14) (n=16)

 Rest 2(1-4) 2(1-4) 4(3-5) 0.06
 Movement 6(4-8) 4.5(2-6) 7.5 (5.5-8.5) 0.02
 Maximum 6(4-8) 5.5(4-8) 9(8-10) 0.002
 Average 4 (3-5) 4.5(2-5) 4(4-6.5) 0.53
SAOAE 2(1-4) 2(1-3) 3(2-4.5) 0.14
Morphine (mg) 1.5 (0-12) 0 (0-15) 4(0-21.5) 0.39
Oxycodone (mg) 20 (10-40) 20 (15-30) 45 (35-60) 0.02 *

LD=low-dose, HD=high-dose, VNRS=verbal numerical rating
scale for pain expressed as median (IQR), SAOAE=subjective
assessment of analgesic efficacy as median (IQR). Drug doses
are median (IQR). P values are from Kruskal-Wallis rank tests. *
Ordinal logistic regression adjusted for age.

Postoperative data from 48 to 72 hours and follow-up

 Control Treatment P value
 (n=28) (n=28)
 Rest 2 (1-3) 1 (0-3) 0.21
 Movement 5 (3-5) 4 (2-5) 0.49
 Maximum 5 (4-6) 5 (3.5-7) 0.85
 Average 3 (2-5) 3.5 (2-5) 0.95
SAOAE 1.5 (0-3) 2 (1-3) 0.18
Morphine (mg) 0 (0-5) 0 (0-0) 0.11
Oxycodone (mg) 15 (10-30) 25 (15-40) 0.04
Length of stay (days) 7 (6-7) 7 (5-8) 0.61
Complications 2 * 4 ([dagger]) 0.67

VNRS=verbal numerical rating scale for pain expressed as median
(IQR), SAOAE=subjective assessment of analgesic efficacy as
median (IQR). Drug doses are median (IQR). Lengths of stay are
median (IQR). All P values are from Wilcoxon rank sum tests.

* Prolonged wound drainage; deep venous thrombosis,
([dagger]) Myocardial infarction; pulmonary embolus; prolonged
wound drainage and infected knee; infected knee.
COPYRIGHT 2009 Australian Society of Anaesthetists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Reeves, M.; Skinner, M.W.
Publication:Anaesthesia and Intensive Care
Article Type:Report
Geographic Code:8AUST
Date:Nov 1, 2009
Previous Article:Derivation and prospective testing of a two-step sevoflurane-[O.sub.2]-[N.sub.2]O low fresh gas flow sequence.
Next Article:Comparison of equipotent doses of ropivacaine-fentanyl and bupivacaine-fentanyl in spinal anaesthesia for lower abdominal surgery.

Related Articles
Biphasic finite element modelling of contact mechanics of hemi-arthroplasty of human hip joint. Part I: metal on cartilage contact.
Intraoperative ketamine may influence persistent pain following knee arthroplasty under combined general and spinal anaesthesia: a pilot study.
Comparison of equipotent doses of ropivacaine-fentanyl and bupivacaine-fentanyl in spinal anaesthesia for lower abdominal surgery.
Repeated intrathecal administration of ropivacaine causes neurotoxicity in rats.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters