Skip to main content

Table 2 Randomised controlled trials of local anaesthetic infiltration in total knee and hip replacement

From: Local anaesthetic infiltration for peri-operative pain control in total hip and knee replacement: systematic review and meta-analyses of short- and long-term effectiveness

Study country date

Inclusion patients (intervention: control)

Common treatment

Latest post-surgical follow up Outcomes Losses to follow up (intervention/ control) Risk of bias summary

Intervention treatment (infiltrate volume) Further treatment (if given)

Control

TOTAL HIP REPLACEMENT STUDIES

Aguirre et al. 2012 [34] Switzerland Not specified

THR (minimally invasive) N = 76 (38:38) 58:58 years 53:50% female

Spinal anaesthesia, PCA morphine

48 hours and to 3 months Intra-venous morphine consumption, VAS pain at rest and with motion, electrocardiogram, skin inflammation or infection, satisfaction. 4 (2:2) lost to follow up, 3/4 caused by catheter dislocation Low risk of bias

20 ml solution containing 60 mg ropivacaine injected into wound before closure. Further continuous infusion through catheter

20 ml placebo injection of saline. Continuous infusion of saline through catheter

Andersen KV et al. 2007 [16] Denmark 2005–2006

THR, OA, elective N = 80 (40:40) 62:61 years 90:85% female

Spinal, post-operative oral oxycodon hydrochloride as required

96 hours VAS pain, length of stay, time to mobilisation, side effects and complications, motor block (Bromage scale) 5 (2:3) patients lost to follow up Unclear (blinding of outcome assessment)

101.5 ml solution containing 200 mg ropivacaine, 30 mg ketoralac and 0.5 mg epinephrine infiltrated during surgery. Further infiltrate through catheter intra-articularly 8 hours after surgery.

Epidural infusion of ropivacaine and morphine

Andersen LJ et al. 2007 [27] Denmark Date not specified

THR, OA, uncemented, >80 years N = 37 (19:18) 62:64 years 84:56% female

Spinal anaesthesia, self-administered oral oxycodone as rescue medication

6 weeks VAS pain at rest and on leg raise up to 8 hours, WOMAC pain to day 4, WOMAC pain, stiffness and function after 1,2,4,6 weeks, EQ5D at 6 weeks, patient controlled analgesic use to discharge, adverse events 3 patients out of 10 not fitting inclusion criteria were identified retrospectively. No losses to follow up Low risk of bias

151.5 ml saline solution containing 300 mg ropivacaine, 30 mg ketorolac, and 0.5 mg adrenaline infiltrated during surgery Further infusion through catheter on day 1.

Saline placebo infiltration Saline placebo infused through catheter on day 1

Bianconi et al. 2003 [10] Italy Date not specified

THR and TKR (78% THR), elective N = 37 (18:19) 66:64 years 79:83% female

Spinal anaesthesia. Loading dose of intravenous morphine at end of surgery

72 hours VAS pain at 2,4,8,12,24,48,72 hours, opioid consumption (rescue medication), adverse events, length of hospital stay, patient satisfaction No losses to follow up Low risk of bias

40 ml saline containing 200 mg ropivacaine infiltrated at end of surgery. Further ropivacaine infusion through catheter for 55 hours after closure. Intravenous saline infusion for 24 h after surgery.

No placebo infiltration during surgery. Saline infusion through catheter for 55 h after closure. Intravenous morphine plus ketorolac infusion for 24 h.

Busch et al. 2010 [30] UK 2003–2005

THR, OA, age <80 years N = 64 (32:32) 61:65 years 50:54% female

General or spinal anaesthesia, PCA morphine

2 years VAS at rest and activity, morphine consumption (PCA), VAS satisfaction, complications, Harris Hip Score, WOMAC, length of hospital stay No losses to follow up Low risk of bias

100 ml saline solution containing 400 mg ropivacaine, 30 mg ketorolac, 5 mg epimorphine, and 0.6 ml epinephrine (1:1000) infiltrated during surgery.

No placebo infiltration

Dobie et al. 2012 [35] UK 2006–2007

THR, OA or RA N = 96 (50:46) 67:67 years 38: 52% female

Spinal, general, intravenous morphine after surgery as required.

6 days VAS at 24 h, morphine consumption, walking and stair test, mobilisation velocity and day, sit to stand test, home readiness, hospital stay, Iowa Level of Assistance Scale 4 (4:0) patients did not receive intervention as planned. Intention to treat results. Some data missing for 1 control Low risk of bias

160 ml saline solution containing 200 mg levobupivacaine and adrenaline

No local infiltration

Lee et al. 2009 [29] South Korea 2006–2007 Note: additional pre-emptive analgesia and epidural

THR, 13% OA, 72% Osteonecrosis N = 60 (30:30) 51:55 years 37:43% female

General anaesthesia

5 days VAS pain, ambulation, doses of parenteral analgesia, time to straight leg raise, complications No losses to follow up described Unclear (blinding of outcome assessment)

Pre-emptive analgesia with oral Oxycodone and Celecoxib. Epidural anaesthesia. 90 ml saline solution containing 5 mg morphine, 40 mg methylprednisolone and 6.8 mg ropivacaine infiltrated during surgery. Post-operative oral Oxycodone and paracetamol.

No pre-emptive analgesia No epidural No injection during surgery Post-operative intravenous PCA and oral and injected analgesics as required

Liu et al. 2011 [32] China 2008–2009

THR, OA, ASA I–III, <80 years N = 82 (41:41) 74:74 years 75:77% female

Spinal anaesthesia, PCA morphine

15 days and 9 months (range 6–12 months) for infection Morphine use, VAS pain, surgical outcome, mobilisation (time to straight leg raise and 90 degree flexion) 2 (1:1) lost to follow up Low risk of bias

60 ml saline solution containing 5 mg morphine, 30 mg bupivacaine, 1 ml betamethasone and 0.5 ml epinephrine infiltrated during surgery.

60 ml saline infiltrated during surgery.

Lu et al. 2010 [31] China Not specified

THR, primary N = 40 (20:20) No information on age and sex of patients

No description of common anaesthesia except PCA

48 hours VAS pain, use of PCA pump, adverse drug reactions No losses to follow up apparent Unclear (limited reporting)

COX-2 inhibitor before surgery. 100 ml solution containing 0.15% ropivacaine infiltrated at end of surgery. COX-2 inhibitor after surgery

No COX-2 inhibitor before surgery. 100 ml saline placebo infiltrated at end of surgery. No COX-2 inhibitor after surgery

Lunn et al. 2011 [33] Denmark 2009–2010

THR, >18 years N = 120 (60:60) 67:67 years 55:65% female

Spinal with or without general. Multimodal oral analgesia

8 hours and to discharge VAS pain at rest and during walking and passive hip flexion, Oxycodone consumption, complications No losses to follow up except “pain during walking” with 18 (11:7) lost to follow up Low risk of bias (except pain during activity: possible risk of bias due to large number of patients unable to complete test|)

150 ml saline solution containing 0.2% ropivacaine and 10 μg/ ml epinephrine infiltrated during surgery.

150 ml saline placebo infiltrated during surgery.

Murphy et al. 2012 [36] Ireland 2009–2010

THR, OA N = 91 (45:46) 57:54 years 49:38% female

Spinal, PCA opioid analgesia

72 hours WOMAC Pain, McGill Pain Questionnaire, VAS pain, morphine consumption, complications 13 (6:7) lost to follow up but some analyses used multi-level modelling to handle missing data Low risk of bias

60 ml saline containing 150 mg levobupivacaine infiltrated during surgery.

60 ml saline placebo

Parvataneni et al. Hip 2007 [28] USA 2005–2006

THR, OA N = 71 (35:36) 64:61 years 40:39% female

Spinal anaesthesia with or without FNB

3 months VAS pain, total narcotic dose, functional recovery including time to straight leg raise, side effects of narcotic use, patient satisfaction No losses to follow up reported Low risk of bias

Intra-operative infiltration of 200–400 mg bupivacaine, 4–10 mg morphine sulphate 300 μg epinephrine, 40 mg methylprednisolone acetate, 75 mg cefuroxime and 22 ml saline. Total volume approximately 33 ml.

No infiltration during surgery Post surgical PCA

Rikalainen-Salmi et al. 2012 [37] Finland 2009–2010

THR, OA, ASA I–III N = 60 (30:30) 65:66 years (followed up) 66:61% female (followed up)

Spinal, propofol if required, oxycodone rescue medication.

8 weeks NRS pain at rest and motion, oxycodone consumption, mobilisation, fulfilment of discharge criteria, satisfaction, adverse events and complications 3 (1:2) lost to early follow up. 7 (4:3) lost to long term follow up Low risk of bias

101 ml solution containing 125 mg levobupivacaine, 30 mg ketorolac infiltrated during surgery 21 ml solution containing 100 mg levobupivacaine and 30 mg ketoralac administered through catheter on morning of first post-operative day

Intrathecal morphine No placebo infiltration Sham catheter attached to skin with 21 ml air administered on morning of first post-operative day (not inserted into joint )

TOTAL KNEE REPLACEMENT STUDIES

Affas et al. 2011 [50] Sweden 2007–2008

TKR, 77.5% OA, 22.5% RA, >18 years, ASA I–III, primary. N = 40 (20:20) 67:69 years 45:60% female

Spinal anaesthesia, PCA morphine

24 hours NRS pain intensity at rest and on movement, 24 hour morphine PCA consumption. No losses to follow up. Missing data analysis reported. Unclear risk of bias (blinding of outcome assessment)

110 ml containing approximately 200 mg ropivacaine, 20 mg ketorolac and 0.33 mg epinephrine infiltrated during surgery. Further intra-articular infiltration through catheter after surgery.

Femoral nerve block. Intravenous ketorolac after surgery. No placebo infiltration.

Andersen KV et al. 2010 [44] Denmark 2007–2008

TKR, >18 years N = 49 (24:25) 67:69 years 43:26% female

Spinal anaesthesia, PCA morphine

72 hours and to discharge. Infection to 30 days VAS/ NRS pain, morphine requirement, side effects and complications, time to achieve discharge criteria, length of stay, 9 (3:6) patients lost to follow up Unclear risk of bias (blinding of outcome assessment)

151.5 ml saline solution containing 300 mg ropivacaine, 30 mg ketorolac and 0.5 mg epinephrine infiltrated during surgery. Further continuous infusion through catheter after closure.

Epidural infusion of ropivacaine. Post-operative intravenous ketorolac

Busch et al. 2006 [38] Canada Date not specified

TKR, age <80 years N = 64 (32:32) 66:70 years 50:59% female

General or spinal anaesthesia, PCA morphine

6 weeks VAS at rest and activity, morphine consumption (PCA), VAS satisfaction, complications, Knee Society Score, WOMAC, length of hospital stay No losses to follow up Low risk of bias

100 ml saline solution containing 400 mg ropivacaine, 30 mg ketorolac, 5 mg epimorphine, and 0.6 ml epinephrine (1:1000) infiltrated during surgery.

No placebo infiltration

Carli et al. 2010 [45] Canada 2007–2008

TKR, OA, tricompartmental, cemented. N = 40 (20:20) 71:71 years 75:70% female

Spinal anaesthesia, PCA morphine

6 weeks Morphine consumption, NRS pain at rest and walking, functional capacity, ability to walk 30 m, physical activity, SF-12, WOMAC No losses to follow up Low risk of bias

Solution of ropivacaine (0.2%), 1 ml of ketorolac (30 mg/ml), and 0.5 ml of epinephrine (1 mg/ml) with a total volume of 100 ml infiltrated during surgery. Further infusion through catheter after closure

Continuous femoral nerve block Saline injection Post-surgical infusion of saline

Chen et al. 2012 [52] China 2008

TKR, OA, age <76 years. N = 81 (40:41) 66:65 years 75:78% female

Spinal anaesthesia, PCA morphine

15 days and infection to 6 months Total morphine consumption, VAS pain at rest and motion, time to straight leg raise and 90 degree flexion, adverse events including delayed infection 1 (0:1) patient lost to follow up Low risk of bias

Intra-operative injection of a solution of magnesium sulphate (50 mg/kg) and 190 mg ropivacaine in normal saline to a volume of 100 ml.

Intra-operative intra-articular injection of 100 ml normal saline

Essving et al. 2010 [46] Sweden 2007–2008

TKR, OA, ASA I–III, 20–85 years N = 48 (24:24) 72:70 years 54:54% female

General anaesthesia, PCA morphine

3 months PCA morphine consumption, VAS pain at rest and on knee flexion, time to home readiness, length of hospital stay, surgical outcome, functional outcome tests, Oxford Knee Score, EQ-5D, patient satisfaction, adverse events 1 (0:1) patient lost to follow up Low risk of bias

116 ml saline containing 300 mg ropivacaine, 30 mg ketorolac and 0.5 mg epinephrine infiltrated during surgery. 50 ml saline containing 100 mg ropivacaine infiltrated before closure. Further injection of mixture 21 h after closure.

No placebo injections during surgery. Post-surgical injection of saline at 21 hours.

Essving et al. 2011 [51] Sweden 2009–2010

TKR, OA, ASA I–III, age 40–85 years N = 50 (25:25) 71:71 years 64:60% female

Spinal anaesthesia, PCA morphine

3 months VAS pain, PCA morphine, verbal rating scale of satisfaction, functional tests, time to home readiness, Oxford Knee Score, EQ-5D, adverse events 2 (0:2) patients lost to follow up Low risk of bias

Spinal plus intrathecal saline. Injection during surgery of 400 mg ropivacaine (160 ml), 30 mg ketoralac (1 ml) and 0.5 mg epinephrine (5 ml) Further infiltrate through catheter on day 1 and 2

Spinal plus intrathecal morphine No injection during surgery Post-surgical infusion of saline through catheter

Fu et al. 2009 [42] China 2006–2007

TKR, OA, age <80y N = 80 (40:40) 69:68 years 75:78% female

Spinal anaesthesia, PCA morphine

15 days except ROM 90 days, infection 12 months Morphine consumption, VAS pain at rest and activity, ROM, time to straight leg raise, surgical outcomes, complications. No losses to follow up. Missing data imputation described Low risk of bias

60 ml saline containing 5 mg morphine, 30 mg bupivacaine and 1 ml betamethasone infiltrated during surgery.

60 ml saline infiltrated during surgery

Fu et al. 2010 [47] China 2008–2009

TKR, OA, age < 80 years N = 100 (50:50) 68:67 years 76:80% female

Spinal anaesthesia, PCA morphine

15 days except ROM at 90 days and infection to mean 7.5 months (range 6–9 months) VAS pain, morphine consumption (PCA and intramuscular) , time to straight leg raise and 90 degree flexion, surgical outcomes, adverse reactions No losses to follow up Low risk of bias

Oral COX-2 inhibitor and tramadol 1 day before to 1 month after surgery 50 ml saline containing 5 mg morphine, 150 mg ropivacaine, 0.5 ml adrenaline and 1 ml betamethasone infiltrated during surgery.

Oral placebo 1 day before to 1 month after surgery 50 ml saline placebo infiltrated during surgery

Han et al. 2007 1 and 2 [40] Korea 2005–2006 Note: 2 intervention groups

TKR, primary N = (30:30:30) 69:68:67 years 90:80:90% female

Spinal and epidural anaesthesia, PCA morphine

48 hours Incidence of booster PCA for 24 hours, amount of intra-venous tramadol, VAS pain at rest and exercising, side effects, range of flexion. No losses to follow up reported Low risk of bias

1) 50 ml saline solution containing 300 mg ropivacaine, epinephrine (0.25 ml 1:200,000) and 5 mg morphine injected before wound closure. 2) 50 ml saline solution containing 300 mg ropivacaine and epinephrine (0.25 ml 1:200,000) injected before wound closure.

50 ml saline placebo

Koh et al. 2012 [53] Korea 2008–2009

TKR, OA, unilateral N = 101 (49:52) 70:70 years 89:91% female

FNB, spinal anaesthesia, PCA morphine

7 days VAS pain at rest (day 1) and on movement (days 4 and 7), PCA opioid consumption, use of rescue medication, pain compared with expectations, functional recovery (straight leg raise and flexion), satisfaction, side-effects and complications, length of stay. 14 (4:10) did not receive treatment as planned. Results reported by intention to treat Low risk of bias

50 ml saline containing ropivacaine 300 mg, morphine sulphate 10 mg, ketoralac 30 mg, 0.3 mg epinephrine, cefuroxime 750 mg injected/ infiltrated during surgery.

No placebo infiltration reported

Krenzel et al. 2009 [43] USA 2007–2008

TKR, 96% OA elective. N = 67 (35:32), 1 patient with staged bilateral TKR included twice. 67:65 years 57:72% female

FNB, spinal anaesthesia, PCA fentanyl

24 hours PCA fentanyl consumption, NRS pain, functional tests, time to straight leg raise, ambulation distance, surgical outcomes, adverse events No losses to follow up Low risk of bias

20 ml infiltration of 100 mg ropivacaine during surgery.

20 ml saline placebo infiltrated during surgery

Mahadevan et al. 2012 [54] UK Not specified

TKR, OA or RA, unilateral N = 52 (26:26) 68:67 years 54:58% female

FNB, general anaesthesia, PCA morphine.

48 hours and to discharge VAS pain, morphine consumption, active ROM, length of hospital stay. No losses to follow up reported Low risk of bias

25 ml saline containing 0.375% levobupivacaine infiltrated during surgery.

Sciatic nerve block No placebo infiltration reported

Meftah et al. 2012 [55] USA 2010–2011

TKR, unilateral N = 90 (45:45) 65:67 years 64:64% female

Pre-emptive analgesia

3 days and to discharge. 6 months for infection, fracture and re-operation. Pain at rest and ambulation, readiness for discharge.1 (1:0) lost to all follow up, 6 (4:2) lost to readiness for discharge follow up Unclear (blinding of outcome assessment)

45.1 ml saline solution containing marcaine (400–800 mg, morphine sulphate 8 mg, adrenaline 0.3 mg, antibiotic 750 mg, corticosteroids 40 mg injected during surgery.

FNB. PCA epiduralNo placebo injection reported

Ng et al. 2012 [56] China 2008–2010 Note: crossover design. Patients having both knees replaced

TKR, OA N = 32 (16:16) surgeries but 16 patients only having 2 TKRs 3 months apart. 70:70 years 88:88% female

General anaesthesia, remifentanil infusion, PCA morphine

3 days and to discharge Pain score at rest and motion, total morphine consumption, Knee Society Score, ROM, quadriceps power, satisfaction, adverse events and complications.No losses to follow up reported Low risk of bias

101.5 ml saline solution containing 300 mg ropivacaine, adrenaline 1 mg and triamcinolone acetonide 40 mg infiltrated during surgery. Femoral catheter inserted and saline infused.

Femoral nerve block. Wound infiltration with 101.5 ml saline.

Parvataneni et al. Knee 2007 [28] USA 2005–2006

TKR, OA N = 60 (31:29) 69:71 years 45:52% female

Spinal anaesthesia with or without FNB

3 months VAS pain, total narcotic dose, functional recovery including time to straight leg raise, side effects of narcotic use, patient satisfaction No losses to follow up reported Low risk of bias

Intra-operative infiltration of 200–400 mg bupivacaine, 4–10 mg morphine sulphate 300 μg epinephrine, 40 mg methylprednisolone acetate, 75 mg cefuroxime and 22 ml saline. Total volume approximately 33 ml.

No infiltration during surgery Femoral nerve block at end of surgery Post surgical PCA Effort to conceal allocation but no sham epidural

Spreng et al. no iv injection 2010[48] Norway 2007–2009

TKR, unilateral, non-cemented, no patella resurfacing, age >17 years, ASA I–III. N = 68 (34:34) 67:66 years 61:67% female

Spinal. Propofol if indicated. PCA morphine

72 hours and to discharge VAS at rest and during knee flexion, morphine consumption, functional recovery, length of stay, satisfaction, mobilisation including walking distance, adverse events2 (1:1) lost to follow up Low risk of bias

150 ml saline solution containing 150 mg ropivacaine, 0.5 mg epinephrine, 30 mg ketorolac and 5 mg morphine infiltrated during surgery. Knee injected through catheter with ropivacaine and ketorolac solution after 22–24 hours Intravenous injection with saline at 22–24 hours

48 hours of epidural analgesia as soon as spinal started to wear off No wound infiltration during surgery. No injections through sham catheter. No sham epidurals

Spreng et al. with iv injection 2010 [48] Norway 2007–2009

TKR, unilateral, non-cemented, no patella resurfacing, age >17 years, ASA I–III. N = 68 (34:34) 67:66 years 61:67% female

Spinal anaesthesia, propofol if indicated, PCA morphine

72 hours and to discharge VAS at rest and during knee flexion, morphine consumption, functional recovery, length of stay, satisfaction, mobilisation including walking distance, adverse events 2 (1:1) lost to follow up Low risk of bias

150 ml saline solution containing 150 mg ropivacaine and 0.5 mg epinephrine infiltrated during surgery. Also intravenous injection of 1 ml ketorolac (30 mg/ml) and 5 ml morphine (1 mg/ml).Knee injected with saline at 22–24 h (catheter) Intravenous injection with ketoralac at 22–24 h

48 hours of epidural analgesia as soon as spinal anaesthetic started to wear off. No wound infiltration during surgery. No injections through sham catheter

Thorsell et al. 2010 [49] Sweden Not specified

TKR, OA or RA N = 85 (46:39) 69:72 years (followed up) 81:73% female (followed up)

Not specified, probable PCA

4 days and to discharge VAS pain, morphine consumption, satisfaction, mobilisation getting out of bed without assistance, walking with crutches), functional recovery, length of hospital stay 21 (13:8) patients lost to follow up data reported Possible bias (large uneven losses to follow up, group allocation by date of birth)

Spinal anaesthesia 156 ml solution with 300 mg ropivacaine, 0.5 mg adrenaline and 30 mg ketorolac infiltrated during surgery. Further infiltrate through catheter intra-articularly on post-operative day 1.

Spinal or epidural analgesia No placebo infiltration reported Post-operative pain relief with ropivacaine infusion through epidural catheter.

Toftdahl et al. 2007 [18] Denmark 2005–2006

TKR, OA with planned spinal anaesthesia 77 (40:37) 70:72 years 63:60% female

Spinal and after surgery immediate release oxycodone and intravenous morphine if required

4 days and to discharge NRS pain, opioid consumption, mobilisation (able to walk >3 metres, able to hold quadriceps tension for > 5 sec), length of hospital stay, adverse events and complications 4 (3:1) patients lost to follow up Unclear (blinding of outcome assessment)

152 ml solution containing 300 mg ropivacaine, 30 mg ketorolac and 0.5 mg epinephrine infiltrated during surgery. Further infiltrate through catheter intra-articularly on day of surgery and post-operative day 1.

Femoral nerve block prior to spinal anaesthesia No placebo infiltration Post-surgical continuous femoral nerve block

Vendittoli et al. 2006 [39] Canada 2003–2004

TKR, 95.2% OA N = 42 (22:20)Ages not specified 73:70% female

Spinal anaesthesia, PCA morphine

5 days and to discharge VAS pain at rest and during physiotherapy exercise, PCA morphine consumption, functional recovery, side effects No losses to follow up described Low risk of bias

160 ml solution containing in total 400 mg ropivacaine, 30 mg ketorolac and 0.5 ml adrenaline (1:1000) infiltrated during surgery. Infiltrate through catheter intra-articularly on day1.

No placebo infiltration

Zhang et al. 2007 [41] China 2006–2007

TKR, unilateral N = 60 (30:30) Overall 68 years 83:80% female

PCA morphine

72 hours VAS pain at rest and activity, functional recovery No losses to follow up described Unclear (blinding of outcome assessment)

60 ml solution containing 0.25% bupivacaine, epinephrine (1:200,000) and 10 mg morphine infiltrated during surgery

No placebo injection

  1. TKR: total knee replacement, THR: total hip replacement , OA: osteoarthritis, RA: rheumatoid arthritis, PCA: patient controlled analgesia, FNB: Femoral Nerve Block, VAS: visual analogue scale, NRS: numerical response scale, ROM: range of motion, WOMAC: Western Ontario and McMaster Universities Arthritis Index.