Regional analgesia for improvement of long‐term functional outcome after elective large joint replacement

Arthur Atchabahian; Gary Schwartz; Charles B Hall; Claudette M Lajam; Michael H Andreae

doi:10.1002/14651858.CD010278.pub2

. 2015 Aug 13;2015(8):CD010278. doi: 10.1002/14651858.CD010278.pub2

Regional analgesia for improvement of long‐term functional outcome after elective large joint replacement

Arthur Atchabahian ^1,^✉, Gary Schwartz ², Charles B Hall ³, Claudette M Lajam ⁴, Michael H Andreae ⁵

Editor: Cochrane Anaesthesia Group

PMCID: PMC4566967 NIHMSID: NIHMS717717 PMID: 26269416

Abstract

Background

Regional analgesia is more effective than conventional analgesia for controlling pain and may facilitate rehabilitation after large joint replacement in the short term. It remains unclear if regional anaesthesia improves functional outcomes after joint replacement beyond three months after surgery.

Objectives

To assess the effects of regional anaesthesia and analgesia on long‐term functional outcomes 3, 6 and 12 months after elective major joint (knee, shoulder and hip) replacement surgery.

Search methods

We performed an electronic search of several databases (CENTRAL, MEDLINE, EMBASE, CINAHL), and handsearched reference lists and conference abstracts. We updated our search in June 2015.

Selection criteria

We included randomized controlled trials (RCTs) comparing regional analgesia versus conventional analgesia in patients undergoing total shoulder, hip or knee replacement. We included studies that reported a functional outcome with a follow‐up of at least three months after surgery.

Data collection and analysis

We used standard methodological procedures expected by Cochrane. We contacted study authors for additional information.

Main results

We included six studies with 350 participants followed for at least three months. All of these studies enrolled participants undergoing total knee replacement. Studies were at least partially blinded. Three studies had a high risk of performance bias and one a high risk of attrition bias, but the risk of bias was otherwise unclear or low.

Only one study assessed joint function using a global score. Due to heterogeneity in outcome and reporting, we could only pool three out of six RCTs, with range of motion assessed at three months after surgery used as a surrogate for joint function. All studies had a high risk of detection bias. Using the random‐effects model, there was no statistically significant difference between the experimental and control groups (mean difference 3.99 degrees, 95% confidence interval (CI) − 2.23 to 10.21; P value = 0.21, 3 studies, 140 participants, very low quality evidence).

We did not perform further analyses because immediate adverse effects were not part of the explicit outcomes of any of these typically small studies, and long‐term adverse events after regional anaesthesia are rare.

None of the included studies elicited or reported long‐term adverse effects like persistent nerve damage.

Authors' conclusions

More high‐quality studies are needed to establish the effects of regional analgesia on function after major joint replacement, as well as on the risk of adverse events (falls).

Plain language summary

Pain control using local anaesthetics to improve surgical results after shoulder, hip and knee replacement surgery

Review question

Doctors can use regional analgesia (injection of local anaesthetics near the nerves or the surgical site) rather than conventional pain control after surgery. Does this choice improve long‐term function after elective major joint (knee, shoulder and hip) replacement? We conducted this systematic review to explore controversy about the use of regional analgesia amidst efforts to limit healthcare costs and demonstrate value for interventions.

Background

People undergo knee, shoulder and hip joint replacement to improve their mobility, pain and function. Local anaesthetics are injected near the nerves or the surgical site to decrease pain after surgery. Regional analgesia has been shown to lead to better surgical results (joint range of motion, strength, ability to walk, climb stairs, etc.) in the days immediately after surgery. It is unclear whether this effect lasts beyond three months.

Study characteristics

We searched electronic databases and abstracts for relevant studies. The search was last updated in June 2015. We found six studies, with 350 participants followed‐up for at least three months. All of these studies enrolled participants undergoing knee replacement.

Key results

Because the studies did not report the same outcomes, it was difficult to pool the results. We could only pool three out of six randomized controlled trials studying regional anaesthesia for better function after total knee replacement. Pooling data from 140 participants, with range of motion assessed at three months after surgery, there was no statistically significant difference between the use of regional analgesia and conventional pain control with intravenous medications.

None of the included studies examined long‐term adverse effects like persistent nerve damage.

Quality of the evidence

We deemed the quality of the evidence to be very low, as the included studies were not considered to be of high calibre and included few participants.

Conclusions

We do not have enough information to determine if regional anaesthesia improves function after major joint replacement or not. We need more and better clinical trials to decide whether there is an effect of regional analgesia on the surgical results after total shoulder, hip or knee replacement surgery. We need more research to understand if regional analgesia increases the risk of falls after joint replacement.

Summary of findings

Background

Description of the condition

Total joint replacement is the definitive treatment for joint‐destroying conditions such as osteoarthritis and rheumatoid arthritis. Replacement arthroplasty or joint replacement surgery is an orthopaedic surgical procedure that replaces an arthritic or dysfunctional joint surface with an artificial prosthesis. Joint replacement is offered when conservative therapy does not alleviate severe pain or dysfunction of the joint. The number of total replacements of large joints (shoulder, hip or knee) has increased dramatically over the past two decades and will continue to increase as the population ages. However, patients can experience severe postoperative pain that can impede rehabilitation.

Description of the intervention

Provision of regional analgesia, either by a continuous or single‐injection technique, controls pain by blocking the conduction of pain impulses from the periphery to the central nervous system. It has been hypothesized that regional anaesthesia blocks afferent pain impulses that cause muscle spasms during postsurgical ambulation or when using the extremity after surgery. This may be why regional analgesia allows for improved short‐ and long‐term knee flexibility and mobility (Singelyn 1998). The control intervention is a combination of non‐steroidal anti‐inflammatory drugs (NSAIDs) and opioids. NSAIDs reduce swelling and inflammation in the operative site and reduce pain perception in the central nervous system, but used alone, they are sometimes insufficient for controlling pain. NSAIDs also have significant side effects (gastrointestinal bleeding, renal injury) and are often contraindicated, leaving physicians to rely on opioids alone. Opioids act on opioid receptors in the central nervous system and are potent analgesics but have significant side effects including respiratory depression, constipation, nausea and vomiting, which limit their use and patient compliance.

How the intervention might work

Decreasing pain after surgery might facilitate rehabilitation and lead to earlier return of motion with better range of motion and motor power. Improved pain control allows for better patient participation during rehabilitation, for example in range of motion exercises. Early mobilization is probably one of the factors determining long‐term joint function, the outcome of interest. Some studies have shown that regional analgesia can increase passive knee flexion for a few months postoperatively (Capdevila 1999). This should allow for a more productive rehabilitation programme. Increased postoperative analgesia allows for earlier rehabilitation, which in turn should decrease the negative effects of immobilization on muscles and joints. This may help prevent long‐term stiffness of the joint and functional disability (Akeson 1987).

Why it is important to do this review

We are conducting this systematic review to explore controversy about the use of regional analgesia amidst efforts to limit healthcare costs and demonstrate value for each intervention. It is unclear if regional techniques improve long‐term function enough to justify the extra cost and effort involved.

There have been a number of controlled clinical trials evaluating regional analgesia for long‐term functionality after major joint surgery, with often conflicting results. Individual randomized controlled trials (RCTs) point to an improvement of both short‐ and long‐term function with regional anaesthesia. Some studies appear to contradict these findings, making it unclear whether the benefits extend beyond the immediate postoperative period in terms of improved surgical outcome. Decreasing pain after major joint surgery has been shown to improve range of motion in the short term (Capdevila 1999; Ilfeld 2006; Ilfeld 2011; Macfarlane 2009), but it is unclear whether these positive effects persist beyond a few months.

Regional anaesthesia and analgesia are, however, associated with additional expenses—in equipment, time and personnel. They might also impede rehabilitation because of the sensory and motor block they cause and, in the case of lower extremities, increase the risk of falls (Feibel 2009; Ilfeld 2010).

The above points suggest that there is enough evidence to warrant a Cochrane systematic review on the effect of regional anaesthesia for total joint replacement surgery, with important implications in terms of justifying the extra cost involved in providing extended postoperative regional analgesia after total joint replacement. So far, this evidence has not been synthesized.

Individual studies, especially if they contradict each other, fail to inform practitioners and patients and are unhelpful for healthcare decision makers. If we can show that regional anaesthesia improves long‐term functional outcomes based on the pooled data of several RCTs, this evidence might alter clinical anaesthesia practice with a resultant improvement in long‐term functional outcomes after major joint surgery in the ageing population.

While our preliminary search found some studies on long‐term functional outcomes after total knee replacement, we could not identify studies after shoulder or hip replacement with long‐term follow up. Our review may expose a lack of research on regional anaesthesia for long‐term function after shoulder and hip surgery for total joint replacement.

Objectives

To assess the effects of regional anaesthesia and analgesia on long‐term functional outcomes 3, 6 and 12 months after elective major joint (knee, shoulder and hip) replacement surgery.

Methods

Criteria for considering studies for this review

Types of studies

We included studies with a randomized controlled study design that were single‐blinded (outcome assessor) trials. Because regional anaesthesia causes numbness of the affected body part, neither participant nor anaesthesia provider could be reliably blinded to the intervention.

We excluded studies that were not randomized. We initially aimed to exclude studies in which the outcome assessor (that is, the nurse or surgeon evaluating the joint function 3, 6 or 12 months after the surgery) was not blinded, but given the paucity of studies, we ended up including studies in which there was no clear statement as to whether the outcome assessor was blinded to group allocation. This is clearly stated in the 'Risk of bias' tables for each study.

Types of participants

We included studies on participants undergoing elective total joint replacement surgery of the knee (TKR), hip (THR) or shoulder (TSR).

We excluded studies including children or participants undergoing revision arthoplasty.

Types of interventions

We included studies that compared continuous or single‐injection postoperative regional analgesia with placebo. We did not include studies comparing one regional technique versus another.

Regional analgesia included epidural anaesthesia, plexus or nerve blocks, or wound or joint infiltration. This was either the primary comparison in the study or a comparison between subgroups. Studies were included without regard to the type of anaesthesia used (general or regional) for the surgery itself. Pharmacological multimodal analgesia (non‐steroidal anti‐inflammatory drug (NSAID), paracetamol, gabapentin or pregabalin, ketamine, etc.) may or may not have been used, but they were comparable in the experimental and control groups.

Types of outcome measures

We only included studies that reported some assessment of the functional outcome of the operated joint for at least three months following surgery.

Primary outcomes

Our primary outcome was the improvement in joint function 3, 6 or 12 months after joint replacement.

There are outcome scores specific for each joint. We planned to use the appropriate score for each joint: The most commonly used scores are the Knee Society Score following total knee arthroplasty (Knee Society), the Harris Hip Score following total hip arthroplasty (Harris 1969), and the Constant Score following total shoulder arthroplasty (Constant 1987), but only one study used these indicators, and we could not obtain the information needed to score the included participants from the published data or the authors. Each joint has its specificities (surgical and anaesthetic techniques, outcomes), so we did not consider pooling data from knee, hip and shoulder operations. The outcome should have been the difference between the preoperative and the postoperative score, that is, the improvement in function.

Only one study reported a global function score. Three studies, however, reported the knee range of motion, or maximal flexion, and that was also used as a surrogate outcome for knee function.

Secondary outcomes

Quality of life: appropriate score such as the Short Form 36 Health Survey (SF‐36) or the Western Ontario and McMaster Universities Arthritis Index (WOMAC; Woolacott 2012), based on the data available in the articles and from the author
Immediate complications, as well as complications at 3, 6 and 12 months. Complications were to be split into:
1. medico‐surgical (myocardial infarction (MI), congestive heart failure (CHF), infection, pneumonia, etc.); and
2. block‐related (falls, local anaesthetic (LA) toxicity, localized infection, persistent neuropathy, etc.).

However, none of the studies reported long‐term complications in a systematic fashion. In fact, several studies did not even report why participants were lost to follow‐up.

Persistent pain upon movement may also be an important functional outcome after joint replacement surgery. Another Cochrane review (Andreae 2012) was recently published on regional anaesthesia for chronic pain after surgery.

Search methods for identification of studies

We performed an electronic search of commonly used databases and handsearched reference lists of relevant studies as well as conference abstracts.

Electronic searches

In order to be relevant to the current practice, we limited our search to 15 years prior to our first round of searching (in November 2012). We searched the most recent issue of the Cochrane Central Registrar of Controlled Trials (CENTRAL; 2015 Issue 6; see Appendix 1 for the detailed search strategy); MEDLINE on OvidSP (January 1997 to June 2015, see Appendix 2); EMBASE on Ovid SP (January 1997 to June 2015, see Appendix 3); and CINAHL on EBSCOhost (January 1997 to June 2015, see Appendix 4). We combined a free text search with a controlled vocabulary search and imposed no language restrictions.

Searching other resources

We searched the reference lists of identified relevant studies for additional citations. We searched conference abstracts (2011 to 2014) of the American Society of Anesthesiologists, the American Society of Regional Anesthesia and Pain Medicine, the European Society of Anaesthesiology and the European Society of Regional Anaesthesia & Pain Therapy.

We also searched two trial registries (http://www.controlledtrials.com and http://www.clinicaltrials.gov) for relevant studies. The last search was conducted in June 2015.

Data collection and analysis

We present a diagram illustrating the process of the searches and selection (Figure 1), and we followed the recommendations of the QUORUM and PRISMA statements (Moher 1999; Moher 2010).

Selection of studies

Two review authors (AA and GS) screened the abstracts of all publications obtained from the results of the above searches. We independently screened all titles and abstracts for eligibility. For trials that appeared to be eligible RCTs, we obtained and inspected the full text to assess their relevance based on preplanned criteria for inclusion. For the full‐text studies reviewed, we noted the reasons for study exclusion (See Excluded studies).

In the case of disagreement between AA and GS, a third author (MHA) acted as a referee. We reached decisions by consensus, with MHA casting the decisive vote if needed. If necessary information was not available in the published article, we contacted the corresponding author to request it.

Data extraction and management

We developed a standard data collection form (Appendix 5). We recorded details of trial design, preoperative assessment, interventions and outcome measures. We performed a pilot run and revised the data sheet accordingly. For each study, two authors (AA and GS) independently extracted information and entered the data in Review Manager (RevMan 5.3).

We extracted the composite function score used as the primary outcome data for the joint being studied, as well as the range of motion when available.

The following secondary outcomes were extracted, when provided.

Quality of life (appropriate score such as the SF‐36 or the WOMAC, based on the data available in the articles and from the authors).
Complications (medico‐surgical and block‐related) at 3, 6 and 12 months.

Two authors (AA and GS) independently extracted, collected and recorded data in the data extraction form. We resolved any discrepancies between the data extracted through discussion. If we were unable to reach a consensus, we consulted a third author (MHA). If information on data or trial methodology was not reported in the published articles, GS or AA contacted the corresponding author of the relevant trial to solicit the missing information.

Assessment of risk of bias in included studies

Two review authors (AA, GS) independently evaluated each report that met the inclusion criteria. We contacted authors for missing information regarding their methods. We graded study quality on the basis of a checklist of design components: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and other bias. We accepted single‐blinding as the obvious effects of regional anaesthesia made blinding of participants and caregivers all but impossible. We initially insisted on outcome observer blinding, but given the paucity of studies, we ended up including studies in which there was no clear statement as to whether the outcome assessor was blinded to group allocation.

We achieved consensus by informal discussion. A third author (MHA) served as final arbiter when needed.

We considered a trial as having a low risk of bias if we concluded that all domains were adequate. We considered a trial as having a high risk of bias if one or more domain was inadequate or unclear. We listed all omitted studies and detailed the reason for each exclusion (Moher 1999).

We reported the 'Risk of bias' table as part of the Characteristics of included studies table.

Measures of treatment effect

Joint function was measured with continuous outcome scores. We endeavoured to use the appropriate outcome score specific for each joint (for example, the Knee Society Score following total knee arthroplasty), clearly stating the minimal clinically significant difference for the score used for each joint in the review.

We evaluated primary and secondary outcomes separately for each joint (hip, knee, shoulder), feeling that the anatomical and physiological differences between shoulder, hip and knee surgery argue against pooling or comparing studies on different joints (Higgins 2011).

We expected that studies would not all report outcomes using the same scale, even if they focused on the same joint. In this case, we planned to employ the standardized mean difference as our summary statistic. As discussed in chapter 9 of the Cochrane Handbook for Systematic Reviews of Interventions, these outcome scales, while not actual continuous outcomes, can be reviewed using the same statistical methods (Higgins 2011). The use of different scales does not reflect real differences in variability among study populations but rather subjective preferences by the study authors. For studies that did not report a continuous score, we attempted to obtain the information needed to score the included participants, either by extracting it from the published data or soliciting it from the authors. However, available data were insufficient to do so.

For the secondary outcomes, lack of data also precluded any meaningful analysis.

Quality of life was also to be expressed as a score; as discussed above, the standardized mean difference would have been used if studies used different scales.
Complications and adverse events (medico‐surgical and block‐related) are dichotomous events, and we would have used the risk ratio as a measure to compare experimental group versus control group risk.

We would have used the mean difference or the standardized mean difference, as appropriate.

Unit of analysis issues

We intended to use the participant as our unit of analysis rather than each operated joint. If studies included participants operated on bilaterally, for example the left and the right knee on the same day, we would have used (and would have expected the authors to report) an average of the joint scores of both knees. However, no such studies were included.

We excluded (or given the nature of the condition and intervention studied we did not expect to find) cluster‐randomized trials, cross‐over trials, or individuals undergoing more than one intervention. In the case of studies with multiple treatment arms, we pooled the groups that used different techniques of regional analgesia and the groups that did not use regional analgesia. In the case that there were multiple observations at different time points for the same outcome, we chose the one closest to the interval being studied.

Dealing with missing data

We contacted the corresponding authors of included studies to obtain any missing information and clarify data inconsistencies. If we encountered significant skew in the outcome data, we solicited individual patient data from the primary authors. Unfortunately, we could obtain additional data only from one author (Nader 2012).

We conducted an intention‐to‐treat (ITT) analysis.

Assessment of heterogeneity

We addressed heterogeneity between studies by stratifying studies based on the joint replaced, as described above (see Measures of treatment effect). In addition, we grouped studies based on differences in rehabilitation and the type of regional analgesia interventions, for example, continuous versus single injection peripheral nerve blocks, or epidural analgesia versus peripheral nerve blocks, as discussed below (see Subgroup analysis and investigation of heterogeneity).

We investigated study heterogeneity between studies on each joint using a Chi² test and calculation of the I² statistic (Higgins 2002). However, we used the random‐effects model regardless of the I² statistic for the reasons discussed (see Data synthesis).

Assessment of reporting biases

We contacted authors to request missing data. We countered time lag bias by repeating our search just prior to the submission of our work. To prevent language bias, we did not impose a language restriction.

We could not examine publication bias using graphical and statistical tests (funnel plot, Egger’s test), as planned, as there were fewer than 10 studies in each subgroup.

Data synthesis

We intended to calculate the standardized mean difference (SMD) of the various continuous scores between the groups stratified for each joint, as described in chapter 9 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). If we found skewed data, we planned to use individual patient data for our evidence synthesis, but we were unable to procure the data from the authors. Hence, we did not perform a meta‐analysis of our primary long‐term functional outcome.

We planned to pool the standardized mean differences using the random‐effects model for meta‐analysis, which would better incorporate the clinical heterogeneity typical among small studies investigating the long‐term effects of regional anaesthesia (Andreae 2012).

We chose the random‐effects model because we wanted to address the question, what is the average intervention effect? (Higgins 2011). We assumed that the different studies were estimating randomly different yet related intervention effects (Higgins 2011). By choosing the more conservative random‐effects model, confidence intervals for the average intervention effect would be wider. We anticipated between‐study heterogeneity (including but not limited to subtle variance in regional anaesthesia delivery) and preferred the random‐effects model because of its more cautious estimate of any treatment effect (DerSimonian 1986).

We extracted dichotomous data on adverse effects, yet these were typically reported inconsistently and anecdotally. We pooled the adverse effects (falls) in the studies that reported that outcome.

Subgroup analysis and investigation of heterogeneity

We performed subgroup analysis on several potential sources of heterogeneity: choice of anaesthesia for the joint replacement proper, adjuvant medications and multimodal pain therapy, differences in rehabilitation, exclusion criteria and comorbidities.

We accepted different choices of local anaesthetics as comparable treatments because we considered the pharmacodynamic differences between the typical local anaesthetics used to be negligible. Likewise, the pharmacokinetic differences are obliterated by the use of continuous blocks. However, the duration of action of a single‐shot block differs substantially from a continuous infusion, just as an epidural technique differs from a plexus or nerve block technique; hence we considered a subgroup analysis to investigate the differences between single‐injection versus continuous analgesia interventions, and between epidural versus plexus or nerve block techniques.

We investigated the sensitivity of our effect measure to the exclusion of studies that did not have similar pain control schemes in both groups. An example would be studies in which an NSAID was administered locally in the experimental group, along with a local anaesthetic, while the control group did not receive NSAIDs by any route.

We scrutinized the forest plot for between‐study heterogeneity and considered the differences in rehabilitation that we were able to extract as potential sources of heterogeneity. When we found both pragmatic and explanatory trials, we excluded the explanatory trials to test if the pragmatic trials alone still show an effect in spite of the inclusion of a wider range of participants and hence possibly larger standard deviations.

Sensitivity analysis

We planned to carry out a sensitivity analysis on the quality characteristics of the included studies, primarily reflected by our judgments described in the risk of bias tables, to explore the influence of the high risk of bias studies on the overall pooled result. However, we did not perform this analysis because there were not enough studies, and the risk of bias was low or unclear.

We investigated whether choosing a fixed‐effect model versus a random‐effects model would have changed the result of our evidence synthesis.

Summary of findings table

We reported the data stratified according to the joint replaced and in subgroups of regional techniques as described above. We presented the data in forest plots for outcomes at 3, 6 and 12 months. Because of the small number of studies and the high heterogeneity in the techniques used and the outcomes reported, we were unable to use the principles of the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) system (Guyatt 2008) to assess the quality of the body of evidence associated with specific outcomes (joint function, quality of life and complications) in our review or to construct a 'Summary of findings' table using the GRADE software; the 'Summary of findings' table was constructed manually. The GRADE approach appraises the quality of a body of evidence based on the extent to which one can be confident that an estimate of effect or association reflects the item being assessed. The quality of a body of evidence considers within‐study risk of bias (methodologic quality), the directness of the evidence, heterogeneity of the data, precision of effect estimates and risk of publication bias.

The Table 1 and Table 2 reflects the fact that the evidence is of very low quality and is insufficient to draw conclusions regarding the efficacy of regional analgesia to improve functional outcome following TSR, TKR or THR.

Summary of findings for the main comparison. Regional anaesthesia compared to conventional pain control for total knee replacement patients: joint functional outcome (Knee Society Score compared to preoperative value), 3 and 6 months after surgery.

Regional anaesthesia compared to conventional pain control for total knee replacement patients
Patient or population: total knee replacement patients Settings: university hospital Intervention: regional analgesia Comparison: conventional pain control
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect  (95% CI)	No. of Participants  (studies)	Quality of the evidence  (GRADE)
	Assumed risk	Corresponding risk
	Conventional pain control	Regional analgesia
Knee Society Score compared to preoperative value Follow‐up: 3 months	The mean Knee Society Score compared to preoperative value in the control group was + 42.87 points	The mean Knee Society Score compared to preoperative value in the intervention group was 1.76 points higher	Impossible to estimate as no standard deviation was provided	30 participants  (1 study)	⊕⊝⊝⊝  very low^a,b,c,d,e
Knee Society Score compared to preoperative value Follow‐up: 6 months	The mean Knee Society Score compared to preoperative value in the control group was + 44.70 points	The mean Knee Society Score compared to preoperative value in the intervention group was 4.03 points higher	Impossible to estimate as no standard deviation was provided	30 participants  (1 study)	⊕⊝⊝⊝  very low^a,b,c,d,e
The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  CI: Confidence interval.
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Study Blinding^a	Population: Enrolled/analysed/followed up to at least 3 months (no. per group)	Intervention(s)	Control	Postoperative outcomes
Kadic 2009 Single	58/53/48 (27/26)	CFNB	No block	Knee function after 3 months (knee flexion, Knee Society score and WOMAC pain, stiffness and function subscale)
Nader 2012 Unblinded	62/62/60 (31/31)	CFNB (Epidural infusion until morning of POD 1)	No block (epidural infusion until morning of POD 1)	Knee flexion at 1, 6 and 12 months Functional outcome on POD 1, POD 2, POD 3, 1 month, 6 months, and 12 months
Singelyn 1998 Unblinded	45/45/45 (15/15/15)	CFNB Epidural analgesia	No block	Knee flexion at 6 weeks and 3 months
Tammachote 2013 Single	59/57/57 (28/29)	Periarticular infiltration	Intrathecal morphine	ROM at 2, 6 and 12 weeks Modified Thai version of the WOMAC score (at 6 and 12 weeks)
Wu 2014 Unblinded	79/60/60 (30/30)	CFNB	No block	Knee Society Score (difference with preoperative value) at discharge, at 6 weeks, 3 months and 6 months
Zhang 2011 Double	96/80/80 (26/27/27)	Periarticular infiltration Periarticular infiltration then ropivacaine/ketorolac infusion	Saline intra‐articular infusion	Maximum knee flexion after 7 days (early recovery) and 90 days (late recovery)

Study	Surgical anaesthesia	Group	Analgesia timeline					Analgesic regimen (common to all within‐study groups)
			POD 0	POD 1	POD 2	POD 3	POD 4
Kadic 2009	Spinal	Exp	CFNB			None		IV morphine PCA Diclofenac + paracetamol
		Con	None
Nader 2012	Spinal‐epidural	Exp	CFNB + epidural	CFNB	None			Hydrocodone / paracetamol PRN If needed, extended‐release oxycodone + hydromorphone PRN
		Con	Epidural	None
Singelyn 1998	GA	Exp	Epidural			None		IV morphine PCA IV propacetamol PRN IM piritramide PRN
		Exp	CFNB			None
		Con	IV morphine PCA			None
Tammachote 2013	Spinal	Exp	Periarticular infiltration	None				IV ketorolac PCA until POD 3, then PO naproxen and tramadol PRN PO paracetamol and amitriptyline
		Con	Intrathecal morphine	None
Wu 2014	Spinal	Exp	CFNB			None		Paracetamol + sustained release diclofenac Oral opioids (codeine or morphine) Metoclopramide, senna and famotidine to prevent side effects
		Con	IV morphine PCA			None
Zhang 2011	GA	Exp	Periarticular infiltration	Placebo		None		Celecoxib IV morphine PCA IM morphine PRN
		Exp		Ropivacaine + ketorolac infusion		None
		Con	Placebo			None

Study	Outcomes	Results
Study	Outcomes	Control	Experimental	P value^a
Primary total knee replacement
Kadic 2009	Knee flexion at 3 months	110.0 (range 100.0 to120.0)	110.0 (range 110.0 to 112.5)	NS
	Knee Society Score (knee component) at 3 months	83.2 (± 13.2)	83.8 (± 12.8)	0.87
	Knee Society Score (function component) at 3 months	58.5 (± 21.2)	61.2 (± 29.3)	0.70
	WOMAC (pain) at 3 months	80.0 (± 18.7)	83.8 (± 12.9)	0.39
	WOMAC (stiffness) at 3 months	71.3 (± 22.4)	75.6 (± 17.4)	0.44
	WOMAC (function) at 3 months	71.8 (± 19.5)	80.4 (± 10.5)	0.05
Nader 2012^b	Knee flexion at 6 months	124.47 (± 8.32)	123.31 (± 9.53)	0.61
	Knee flexion at 12 months	124.56 (± 7.67)	124.50 (± 10.37)	0.98
	WOMAC at 6 months	11.04 (± 12.50)	14.18 (± 15.70)	0.39
	WOMAC at 12 months	6.76 (± 7.26)	17.39 (± 17.84)	0.003
Singelyn 1998	Knee flexion at 3 months	116 (± 11)	121 (± 12)	0.24
Tammachote 2013	Knee flexion at 3 months	140 (± 4)	139 (± 4)	0.35
Tammachote 2013	WOMAC at 3 months	9 (± 4)	9 (± 4)	1.00
Wu 2014	Knee Society Score (difference with preoperative value) at 3 months	+ 42.87	+ 44.63	0.78
Wu 2014	Knee Society Score (difference with preoperative value) at 6 months	+ 44.7	+ 48.73	0.51
Zhang 2011	Knee flexion at 3 months	105.7 (3.4)	112.2 (3.8)	P value < 0.001

Study	Adverse events attributable to the regional anaesthesia technique		Common short‐term adverse events not directly related to regional anaesthesia
Study	Control	Experimental	Event	Experimental	Control	P value^a
Kadic 2009	None	None	PONV	2	13	0.0007
			Drowsiness	0	3	0.11
			Constipation	0	4	0.05
			Urine retention	0	1	0.49
Nader 2012	1 fall (Exp); NS	4 DVTs (Con); P = 0.04 3 positive joint aspirates (Exp); P = 0.08	Nausea	0	0	1.00
			Vomiting	0	0	1.00
			Pruritus	2	1	1.00
Singelyn 1998	6 catheter technical issues (Epidural); P < 0.001 vs. CFNB and IV PCA Falls, LA toxicity, etc.: not reported	Not discussed	PONV	9 (5/4)	6	0.52
			Hypotension	1 (0/1)	0	1.00
			Urine retention	6 (0/6)	2	0.70
Tammachote 2013	Not discussed	No wound problem Other complications not discussed	PONV	10	20	0.02
			Pruritus	3	11	0.03
			Urine retention	0	0	1.00
			Respiratory depression	0	0	1.00
Wu 2014	None	5 DVTS (2 Exp, 3 Con) 5 cases of bleeding requiring transfusion (2 Exp, 3 Con) 5 shocks in ward (3 Exp, 2 Con) 2 wound infections (1 Exp, 1 Con)	PONV	8	19	0.009
			Urine retention	3	8	0.18
			Dizziness	5	12	0.08
			Desaturation	0	1	1.00
			Pruritus	not reported
Zhang 2011	None	6 DVTs (2 Exp1, 2 Exp2, 2 Con): NS	PONV	8 (4/4)	7	0.36
			Pruritus	2 (1/1)	1	1.00
			Urine retention	2 (1/1)	2	0.60
			Respiratory depression	1 (0/1)	1	1.00
Pooled 199 in Exp groups 158 in Control groups	not applicable	not applicable	PONV	37/199	67/158	< 0.0001
			Pruritus	7/169	13/128	0.06
			Urine retention	11/199	13/158	0.40

ID #	First author	Journal/Conference Proceedings, etc.	Year	PMID/Identifier unpublished

Intervention	Non‐Regional Comparator	Functionality
Regional/No/Unclear	Yes/No/Unclear	Movement/No/Unclear

Code each paper	ID#	Author(s)	Journal/Conference Proceedings etc	Year	PMID/ Identifier
A		*The paper listed above*
B		*Further papers*
C

Trial characteristics
	Further details
Single centre / multi centre
Country / Countries
Trial design	Randomized Yes No Controlled Yes No Placebo Alternative Treatment Cross‐Over Parallel Prospective Retrospective Groups: 1 2 3 More ‐‐‐‐

Participant characteristics
Further details
Total Hip replacement Total shoulder replacement Total Knee Replacement Other
@ Screening/enrolment: @ Randomization: @ Endpoint:

Age (mean, SD)	Each group all patients
Paediatric Population %	0
Sex of participants (men/women)	Exp Control
Inclusion Criteria
Exclusion Criteria
Comorbidities

Intervention
Regional Nerve Block	Epidural Femoral Block Interscalene other
Active Ingredient	Ropivacaine Bupivacaine Other
Describe the Regional Technique administered

Comparator
Comparator	Placebo General Anesthesia Other Describe:
Comments
Outcomes
Follow up period(s)	Pre First@___ second@___ third@____ fourth@____
Outcome(s)	Dichotomous Continuous Ordinal Responder data Primary: Secondary:

Selection bias/Allocation of intervention
State here method used to generate allocation and reasons for grading	Grade (circle)	Bias likely
Allocation is clearly described. An accepted randomized method is used. Randomization is done at appropriate time point. Details:	Adequate (Random)	No
	Inadequate/ non‐randomized	Yes
	Unclear	Unclear

Performance bias/Concealment of allocation Process used to prevent foreknowledge of group assignment in a RCT, which should be seen as distinct from blinding
State here method used to conceal allocation and reasons for  grading	Grade (circle)	Bias likely?
Concealment of allocation is explained. Provider and participants are unaware of allocation throughout treatment/observation period, respectively. Detail: ___________________________________________________________	Adequate	No
	Inadequate Not pertinent	Yes
	Unclear	Unclear

Detection bias/Blinding		Bias likely
Person responsible for participants care	Yes/No/Unclear
Participant	Yes/No/Unclear
Outcome assessor	Yes/No/Unclear
Other (please specify) _____________________	Yes/No/Unclear
Comments:

Attrition bias/Intention‐to‐treat An intention‐to‐treat analysis is one in which all the participants in a trial are analysed according to the intervention to which they were allocated, whether they received it or not.
Loss to follow up?	Reported reported as none Uncertain/not reported not applicable
ITT Analysis?	Reported not reported no Uncertain not applicable
PP Analysis?	Reported not reported Uncertain not applicable
Were withdrawals described?	Yes No Not clear
How were lost participants/withdraw accounted for:	Last observation carried forward Information collected at end of study Uncertain/not reported Excluded
Comments
Bias likely	Yes/No/Unclear
Selective Reporting Did the study authors report the primary outcomes or is the reported significant outcome a secondary outcome, while the primary outcome of the study was negative.
Primary outcome stated	Prior to study begin, e.g. in protocol After completion of study, e.g. in publication
Outcome was the primary outcome of the reported study	Yes No Unclear
Conflict of Interest and Sponsors Is there a statement on potential conflict of interests or sponsors and did these have an undue influence on the analysis of the data
Conflict of interest	Statement on COI: Yes No
	No indication of relevant COI Unclear Possible conflict of interest apparent
Sponsors	No report on sponsors Unclear information on sponsor Sponsor reported and detailed information
	Possible undue influence apparent Unclear Undue sponsor influence unlikely

Table 1: Responder data @ _ weeks/months	Treatment Group	Comparison Group	Third Group	Total
Number randomized
Number analysed ITT
Number analysed PP

Table 1: Comparison @ _ weeks/months	Treatment Group	Comparison Group	Third Group	Total
Number randomized
Number analysed ITT
Number analysed PP

Table 2: Comparison @ _ weeks/months	Treatment Group	Comparison Group	Third Group	Total
Number randomized
Number analysed ITT
Number analysed PP

Outcome/Instrument	Treatment Group Value (+/‐SD)	Comparison Group Value (+/‐SD)	Between Group Comparison Absolute difference, confidence interval [] and P value

	Treatment Group	Comparison Group
Any adverse event not reported
Withdrawals due to adverse events
Withdrawals due to any reason
Comments:

Code each paper	Author(s)	Journal / Conference Proceedings, etc.	Year	PMID/Identifier	Published
A					Yes/No
B					Yes/No
C					Yes/No

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Knee flexion	3	140	Mean Difference (IV, Random, 95% CI)	3.99 [‐2.23, 10.21]
2 Knee flexion (Infiltration only)	2	110	Mean Difference (IV, Random, 95% CI)	2.76 [‐4.59, 10.11]

Methods	Partially blinded, randomized, controlled trial Random allocation to a group by drawing a sealed envelope from a pile Follow‐up: 3 months
Participants	Subjects: 58 participants enrolled (53 analysed, 48 followed up to 3 months) from one university hospital: Radboud University Nijmegen Medical Centre in the Netherlands Operation: primary total knee replacement 2 groups (of 27 and 26 people) Mean age (SD): group 1, 67.4 (± 12); group 2, 66.8 (± 11) Sex: group 1, 7 males and 20 females; group 2, 7 males and 19 females
Interventions	Standardized anaesthetic: spinal using 15 mg of 5 mg/mL isobaric bupivacaine Group 1 (experimental group): continuous femoral nerve block placed using neurostimulation 5 mL of 20 mg/mL of lidocaine given to test the block with ice Then 20 to 25 mL of ropivacaine 7.5 mg/mL given Postoperatively, an infusion of ropivacaine 2 mg/mL was started at 5 mL/h and adjusted as needed to a maximum of 10 mL/h for the next 48 hours Group 2 (control group): no block Standard analgesic regimen using paracetamol, diclofenac and a morphine IV PCA (bolus 1 mg, lockout 6 minutes)
Outcomes	Pain during the first 48 hours Morphine consumption during the first 48 hours Knee flexion in the first week (days 3 through 6) Knee function after 3 months (knee flexion, Knee Society Score and WOMAC pain, stiffness and function subscale) Catheter position was tested immediately after insertion using lidocaine; however, there is no information as to whether any of the blocks needed additional troubleshooting.
Funding source	Not provided
Declarations of interest	None reported
Notes	—
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"For the randomization procedure, 58 sealed envelopes enclosed a note of either the study or the control group. A blinded operating room nurse drew an envelope, which allocated the patient to one of the groups."
Allocation concealment (selection bias)	Low risk	As above: random draw procedure with sealed envelopes
Blinding of participants and personnel (performance bias)   All outcomes	Unclear risk	No mention of blinding for participants or personnel assessing participants during their hospital stay
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	"The knee function was assessed by a blinded, independent physician, 3 months after surgery at the orthopaedic outpatient centre."
Incomplete outcome data (attrition bias)   All outcomes	Low risk	"[F]ive patients were excluded from the analysis due to protocol violations (one study, two control), a severe adverse reaction to morphine (one control) and for not meeting the inclusion criteria (one study)." At the 3‐month assessment, data from only 48 patients are given (21 in the study group, and 17 in the control group), but no explanation is given as to why 5 more patients were lost to follow‐up.
Selective reporting (reporting bias)	Low risk	Knee Score and WOMAC components reported as mean and SD

PERMALINK

Regional analgesia for improvement of long‐term functional outcome after elective large joint replacement

Arthur Atchabahian

Gary Schwartz

Charles B Hall

Claudette M Lajam

Michael H Andreae

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the condition

Description of the intervention

How the intervention might work

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

1.

Selection of studies

Data extraction and management

Assessment of risk of bias in included studies

Measures of treatment effect

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

Summary of findings table

Summary of findings for the main comparison. Regional anaesthesia compared to conventional pain control for total knee replacement patients: joint functional outcome (Knee Society Score compared to preoperative value), 3 and 6 months after surgery.

Summary of findings 2. Regional anaesthesia compared to conventional pain control for total knee replacement patients: knee flexion 3 months after surgery (as a surrogate for joint functional outcome).

Results

Description of studies

Results of the search

Electronic search

Handsearch

Selection process

Data extraction

Incomplete and raw data

Included studies

1. Overview of included studies.

2. Timeline chart for included studies.

Design

Descriptive characteristics of participants

Intervention

Control

Primary outcomes

3. Summary of reported results.

Secondary outcomes

4. Adverse events.

Excluded studies

Risk of bias in included studies

2.

3.

Allocation

Blinding

Incomplete outcome data

Selective reporting

Other potential sources of bias

Effects of interventions

Primary outcome (Table 3)

Methods	Unblinded, randomized, controlled trial Computer‐generated random allocation sequence Follow‐up: 12 months
Participants	Subjects: 62 participants enrolled (60 followed up to 12 months) from one university hospital: Northwestern University, Chicago, IL Operation: primary total knee replacement 2 groups (of 31 people each) Age: group 1, 65 (IQR 60 to 76); group 2, 64 (IQR 60 to 71) Sex: group 1, 13 males and 18 females; group 2, 7 males and 24 females
Interventions	All participants received spinal‐epidural anaesthesia (10 mg of 5 mg/mL isobaric bupivacaine intrathecally, epidural test dose 3 mL of 15 mg/mL lidocaine with 1:200,000 epinephrine) followed by epidural analgesia (continuous infusion of 3 mL/h of 1mg/mL bupivacaine with hydromorphone 10 mcg/mL, patient boluses of 3 mL with a lockout time of 15 minutes and an hourly maximum of 15 mL) until the morning of POD 1 Group 1 (CFA group): 10 mL bolus of 2.5 mg/mL ropivacaine injected into femoral perineural catheter (placed prior to spinal‐epidural) followed by 5 mL/h infusion of 1 mg/mL ropivacaine until the morning of POD 2; oral analgesic regimen as below for breakthrough pain Group 2 (OOA group): after end of epidural infusion, oral regimen only, as below No adjuvants Oral postoperative analgesic regimen in both groups: hydrocodone 10 mg plus paracetamol 325 mg every 4 to 6 hours for pain as needed; if participants could not achieve a VRSP < 4, the analgesic regimen was changed to sustained release oxycodone 10 mg every 12 hours plus oral hydromorphone 2 mg every 4 hours for breakthrough pain; not different between groups
Outcomes	Primary outcome: knee flexion at 1, 6 and 12 months Secondary outcomes: patient satisfaction during hospital stay, amount of opioid analgesic consumed, verbal rating of pain during hospital stay and follow‐up visits, health‐related quality of life, and functional outcome on POD 1, POD 2, POD 3, and at 1 month, 6 months, and 12 months after surgery Effectiveness of the regional analgesia is not reported
Funding source	Stryker Instruments, Inc. and departmental funds
Declarations of interest	None reported
Notes	—
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Randomization was determined using a computer‐generated random allocation sequence"
Allocation concealment (selection bias)	Low risk	"[G]roup membership was concealed by placing the assignment slip in an opaque envelope that was not opened until after informed consent was obtained."
Blinding of participants and personnel (performance bias)   All outcomes	High risk	No blinding. Participants in the CFA group received a femoral perineural catheter while participants in the OOA group did not.
Blinding of outcome assessment (detection bias)   All outcomes	Unclear risk	No blinding. This is not explicitly stated, but it is possible that the investigators and the surgeon assessing the participants had access to the records that would show whether each participant had received a continuous femoral block.
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Only 1 participant in each group lost to follow‐up after 1 month
Selective reporting (reporting bias)	Unclear risk	The difference in knee flexion on POD 3 and at 1 month is significant but only reported graphically, with no values given. While the components of the WOMAC score are reported as median (range), the scores themselves are not reported.

Study	Reason for exclusion
Affas 2011	Comparison of two regional techniques with no control group; insufficient follow‐up
Allen 1998a	Insufficient follow‐up (2 days)
Allen 1998b	Insufficient follow‐up (24 hours)
Andersen 2008	Insufficient follow‐up (48 hours)
Andersen 2013	Comparison of two regional techniques
Baranović 2011	Insufficient follow‐up (24 hours)
Biboulet 2004	Insufficient follow‐up (24 hours)
Borghi 2002	Insufficient follow‐up (24 hours)
Bozkurt 2009	Insufficient follow‐up (72 hours)
Capdevila 1999	The study included participants who underwent both total knee replacement and knee ligament reconstruction. Despite several requests, we could not obtain the original data to use only the results of the participants having undergone a TKR
Chan 2013	Abstract only. Results could not be obtained from the authors
Chelly 2001	Insufficient follow‐up (5 days)
Chloropoulou 2013	Insufficient follow‐up (24 hours)
Cucereanu Badica 2010	Abstract only. Insufficient follow‐up (48 hours)
Divella 2012	Insufficient follow‐up (3 days)
Hoffmann‐Klefer 2008	Insufficient follow‐up (48 hours)
Ilfeld 2009	All participants received regional analgesia for the first 24 hours, and were then administered placebo or local anaesthetic depending on their group assignment. Thus, there was no real comparison of regional analgesia vs. non‐regional analgesia.
Ilfeld 2011	All participants received regional analgesia for the first 24 hours, and were then administered placebo or local anaesthetic depending on their group assignment. Thus, there was no real comparison of regional analgesia vs. non‐regional analgesia.
Kampe 2001	Insufficient follow‐up (32 hours)
Lang 1999	Letter to the editor commenting on Allen 1998b
Lee 2012	Insufficient follow‐up (48 hours)
Looseley 2013	Insufficient follow‐up (6 weeks). Abstract only
Marino 2009	Insufficient follow‐up (48 hours)
McCarthy 2011	Abstract only, not published as article. Follow‐up of 48 hours, therefore data was not sought out.
Mejia‐Terrazas 2007	Insufficient follow‐up (48 hours)
Mistraletti 2006	Comparison of two regional techniques; insufficient follow‐up
Nigam 2011	Not a regional analgesia technique
Ning 2009	Unable to obtain article; however, from abstract, comparison of 2 regional techniques and insufficient follow‐up (24 hours)
Perrier 2010	Insufficient follow‐up (48 hours); participants undergoing surgery for hip fracture rather than elective joint replacement
Reinhardt 2014	Comparison of intra‐articular infusion of local anaesthetic vs. epidural plus femoral nerve block. Insufficient follow‐up (24 hours)
Singelyn 1999	Non‐randomized; 48‐hour follow‐up
Singelyn 2005	Insufficient follow‐up (48 hours)
Sites 2004	Insufficient follow‐up (24 hours)
Souron 2003	Insufficient follow‐up (48 hours)
Taicher 2010	Abstract only; retrospective review
Tang 2010	Insufficient follow‐up (48 hours)
Tang 2010b	Insufficient follow‐up (48 hours)
Wang 2002	Insufficient follow‐up (until discharge)
Williams 2013	Intra‐articular infusion of local anaesthetic at a low rate (2 mL/h), insufficient to affect postoperative pain scores or morphine use
Wylde 2015	For patients in the total knee replacement group, the comparison was between two regional techniques. Range‐of‐motion data was collected but not published. For patients in the total hip replacement group, only pain and quality‐of‐life data were collected.