Clinical utility of randomized controlled trials of regional anesthesia for lower-extremity surgery: a systematic review

Abstract

Background

Regional anesthesia (RA) is central to multimodal analgesia, yet many supporting randomized controlled trials (RCTs) may lack the rigor needed for evidence-based block selection.

Objectives

To evaluate the clinical usefulness and transparency of RCTs investigating RA interventions for lower extremity surgeries, using the van ’t Hooft et al framework.

Methods

Embase/MEDLINE were searched for RCTs (2014–2024) examining RA interventions for lower-extremity surgery. Data were extracted in duplicate using clinical-utility and transparency criteria. Analyses included descriptive statistics, linear regression for temporal trends, Pearson correlation, and multivariable regression to identify predictors of usefulness.

Results

Among 248 RCTs, patient-centeredness was universal, while feasibility (77.0%) and context placement (77.4%) were common. Pragmatism (2.0%) and value-for-money analyses (0%) were rare. Transparency varied: conflicts of interest were disclosed in 88.7%, pre-registration occurred in 39.9%, public protocols in 10.1%, and raw-data sharing in 1.2%. Transparency improved over time, correlated with clinical-utility scores (r=0.32, p<0.001) and was greater in hospital-funded, self-funded, and higher-impact journals.

Limitations

Usefulness scoring involves subjective judgment; non-English trials were excluded; the absence of formal risk of bias analysis limits interpretability of trial quality.

Discussions/conclusions

Although transparency is improving, most trials remain insufficiently pragmatic, lack economic evaluation, and rarely share data, limiting real-world applicability. The results of this study suggest that aligning funder, journal, and collaborative expectations around open science and pragmatic design could accelerate the generation of high-utility evidence for perioperative block selection.

Introduction

Lower-extremity arthroplasty volumes continue to climb: the American Joint Replacement Registry recorded more than 3.7 million hip and knee replacements between 2012 and 2023, an 18% increase from the previous year’s report.¹ With this increase, regional anesthesia (RA) has become central to perioperative care, with single-injection and continuous peripheral nerve blocks widely adopted and newer techniques such as pericapsular nerve group, supra-inguinal fascia iliaca, and adductor canal with genicular or iPACK supplements showing promise.^2–4 Meta-analytic data already suggest that nerve-block-based anesthesia reduces cognitive dysfunction, respiratory complications, and surgical-site infection after hip and knee arthroplasty, while prospective series show this anesthesia choice as the dominant determinant of ambulatory discharge time.⁵ Yet, anesthesiologists still lack high-quality, patient-oriented evidence to inform block selection, dosing strategies, and catheter management for diverse surgical populations.

Research suggests that much of this evidence gap stems from design and reporting limitations common to RA trials. Many randomized control trials (RCTs) restrict enrollment to relatively healthy patients undergoing primary joint replacement, repeatedly excluding the very cohorts now prevalent in operating rooms: individuals with obesity, diabetic neuropathy, chronic opioid use, or multiple cardiometabolic comorbidities.^6–8 These exclusions may leave anesthesiologists inadequately prepared to tailor blocks for higher-risk populations. Moreover, most trials still favor surrogate or short-term endpoints (block-onset time, intraoperative opioid consumption, or static pain scores within 24 hours) while overlooking outcomes such as time to first unassisted ambulation, engagement in early physical therapy, opioid-free recovery days, length of stay, fall risk, and patient-reported satisfaction with mobility and pain control.^{9 10}

Despite increasing use of RA, high-quality, generalizable evidence remains limited. Trials often exclude patients with common comorbidities and focus on surrogate or short-term outcomes rather than endpoints that matter to recovery and patient experience.^{11 12} Heterogeneity in trial design and incomplete reporting further limit replication and translation into practice. Taken together, these issues mean that many RA RCTs are not practice-informing.

To address these deficiencies, van ’t Hooft et al created a ‘usefulness’ framework that rates RCTs on transparency, feasibility and informational value.¹³ While the van ‘t Hooft framework originally was used for obstetrics, we were able to apply this tool to lower-extremity RA RCTs to assess their readiness for routine anesthetic practice. This framework integrates seven clinical-utility criteria with six transparency criteria to generate a multidimensional assessment of trial relevance. Although the framework has not been formally validated as a measurement instrument, it provides a structured approach to evaluating how well RCTs balance methodological rigor, reporting transparency, and applicability to patient care—elements directly aligned with our study objectives. In contrast to efficacy-focused network meta-analyses and consensus recommendations, our review applies a structured usefulness framework to characterize how well RA trials map to real-world practice and open-science standards. To our knowledge, no prior RA-focused review has operationalized trial usefulness in this manner.

Consistent with our preregistered protocol, we hypothesized that RCTs of RA for lower-extremity surgery published between 2014 and 2024 would show incremental improvement in overall usefulness scores over time. Specifically, we expected transparency practices (eg, pre-registration, funding disclosure, conflict-of-interest statements) to improve more consistently than clinical utility elements (eg, pragmatism, value-for-money analyses), reflecting the stronger influence of recent international and federal mandates emphasizing open science and reproducibility compared with longstanding challenges in pragmatic trial design.

Methods

Reproducibility and study design

This study evaluated the usefulness of published RCTs focused on RA interventions for lower extremity surgery. Studies published between January 1, 2014 and December 31, 2024, were assessed to ensure sample adequacy and capture recent developments. We chose 2014 to capture a decade of RA RCTs reflecting widespread multimodal analgesia and growing transparency practices, while excluding older studies with less relevant design and reporting standards. To ensure transparent and comprehensive reporting, the study adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.¹⁴ In addition, our methodology was informed by recently published guidance in Regional Anesthesia and Pain Medicine on conducting systematic reviews specific to this field. ¹⁵ All study materials are publicly available on our Open Science Framework (OSF) repository.¹⁶

Eligibility criteria/screening

To be included, RCTs must be published in full-text, English, and involve human patients undergoing RA for lower extremity surgery. Only the initial full publication presenting the trial’s primary outcomes was included, while published secondary analyses were excluded. For trials with multiple publications, the earliest and most comprehensive account of the primary results was selected. Trials were excluded if they did not focus primarily on RA interventions for lower extremity surgeries, under the age of 18 years old, were not randomized, focused solely on drug pharmacokinetics/pharmacodynamics, had medication dosage as the main outcome, or were solely safety-focused. Additionally, cohort studies, case series, cross-sectional analyses, case–control studies, review articles, commentaries, editorials, letters, errata/corrigenda, systematic reviews, meta-analyses, pilot studies, protocols, case reports, and unrelated studies were excluded. Two researchers (DW, JS) independently screened all titles/abstracts and full texts against eligibility criteria. Any disagreements were discussed and, if unresolved, were adjudicated by a third researcher (AK) to reach consensus. See online supplemental file for full-text excluded articles and reasoning.

Outcomes

Our review followed a PICO framework: RCTs of adults undergoing RA for lower extremity surgery (Population), evaluating RA techniques (Intervention) against alternative RA, general anesthesia, or placebo/sham (Comparator), with primary outcome usefulness defined by the van ’t Hooft framework (Outcome).

Search strategy

On May 14, 2025, a systematic search was employed in Embase via Elsevier, which simultaneously grants access to MEDLINE (EP). The search identified RCTs focusing on RA interventions for lower extremity surgery. Search filters were adapted from Cochrane and applied to restrict the study by publication date, population, and study design to ensure inclusion of relevant evidence.¹⁷ The complete search strategy is publicly available on OSF.¹⁶ No supplementary sources were consulted. Articles were then compiled into Rayyan, and duplicates were removed by the deduplication tool and screened manually by two independent investigators (DW, JS).¹⁸ The number of records retrieved from each database is reported in figure 1.

Figure 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram for study inclusions. *Article behind paywall or not obtainable through interlibrary loan.

Training

Faculty investigators (MV, AF) trained all team members on study objectives, methodology, inclusion/exclusion, van ’t Hooft framework, and data variables.¹³ Screening was done in Rayyan, with data collected via a pilot-tested Google Form. To ensure consistent coding and uniform decision-making for full data extraction, a written data dictionary was referenced.

Data extraction

For each eligible study, investigators (DW, JS) collected study characteristics and usefulness criteria in a masked fashion using a pilot-tested Google Form. Recorded characteristics included: publication title, PubMed Identifier, journal, impact factor, publication date, registry number, trial continent(s) affiliation(s), intervention type, sample size, funding source(s), and trial phase. The 13-item framework developed by van ’t Hooft et al was used to assess trial usefulness.^{13 19} We selected the van ’t Hooft et al framework because it uniquely integrates clinical utility, pragmatism, and transparency and has been applied across obstetrics, pediatrics, and orthopedics.^{13 19 20} Each criterion was assigned a value of 0 (criterion absent), 1 (partially fulfilled), or 2 (fully fulfilled), for a maximum of 26. Definitions are provided in table 1.

Table 1. Usefulness criteria item descriptions.

Clinical utility item	Description
Problem base	Health impact depends on prevalence, individual burden, and cost, including common and rare diseases
Context placement	Clinical research is most useful when informed by systematic reviews that integrate new findings with existing evidence
Information gain	Informative clinical research requires adequately powered studies focused on critical outcomes, avoiding surrogate measures, and integrating results with prior evidence for meaningful practice changes
Pragmatism	Pragmatic research evaluates the real-world effectiveness of interventions, emphasizing generalizability and practicality over controlled, ideal conditions
Patient-centeredness	Patient-centered research prioritizes questions and outcomes that matter most to patients, involving them in setting priorities and developing core outcomes
Value for money	Value-for-money research evaluates the expected benefits of a study against its costs, using analyses like value-of-information and budget impact to guide design and funding decisions
Feasibility	Feasibility in research depends on realistic recruitment goals, adequate power, and collaborative networks to overcome challenges like patient availability and overestimated sample sizes
Transparency item	Description
8A. Pre-registration	Pre-registration is the process of registering a study in an official registry database before recruiting the first patient
8B. Public protocol	The protocol and data analysis plan should be publicly available before the trial starts and before data analysis
8C. Protocol adherence	Any deviations from the original protocol should be justified, updated in the protocol, and reapproved by the ethics committee, with modification statements included in the manuscript
8D. Funding statement	Funding sources for trial conduction should be clearly stated
8E. Conflict of interest statement	Conflicts of interest should be clearly and completely disclosed
8F. Data availability	Freely available raw data, including statistical code and output, is becoming the norm to ensure transparency, facilitate meta-analyses, and address concerns about study trustworthiness

Patient-centeredness was satisfied if a trial reported ≥1 outcome relevant to perioperative experience, such as pain, nausea/vomiting, time to ambulation, mobility, functional recovery, or patient satisfaction, while pragmatism required meeting multiple design-related elements.

We defined clinical utility as the extent to which an RCT offers information that is relevant, generalizable, and applicable for real-world clinical practice. Transparency was defined as the degree to which an RCT’s design, execution, and reporting are open, complete, and reproducible. Together, clinical utility and transparency can comprise the overall usefulness of a trial.

A guide was made based on previous studies to reduce ambiguity in this evaluation system, which can be found in online supplemental file 4.²⁰ Investigators (DW, JS) extracted data from the first five RCTs for training, resolving discrepancies through discussion. During full data extraction, all items were assessed independently and in duplicate, except the ‘problem base’ item, which clinical investigators (BD, WR) evaluated due to the need for clinical interpretation of trials addressing prevalent real-world health conditions. Another investigator (AK) was available for arbitration. The same investigators then completed data extraction for the remaining included RCTs.

Data analysis

Descriptive analyses were conducted using R (V.4.3.1) in RStudio (V.2023.09.1+494) to summarize frequencies and percentages of trial characteristics. A simple linear regression model was used to examine associations between the publication date and the summed value for each subset. An additional model was applied to analyze the relationship between the publication date and the overall value. Further, the association between the transparency and clinical utility values was visualized and quantified using a Pearson correlation coefficient (p value <0.05). All predictor variables were specified a priori in the preregistered protocol, and no post hoc variables were introduced. Finally, a multivariable linear regression model was used to identify associations between overall usefulness value and the following key predictors: sample size, funding type, trial location, journal type, and impact factor.²¹

Protocol deviations

There were a few deviations from the preregistered protocol. First, the planned risk of bias assessment using the Cochrane RoB 2.0 tool—alongside ChatGPT support was omitted, as ChatGPT was not purpose-built for this task, and RobotReviewer was no longer operational. Second, the “problem base” criterion was evaluated by two resident investigators (BD, WR) independently, rather than in duplicate, due to this criterion’s need for clinical and contextual expertise. Finally, single-group assignment trials were included to avoid excluding a small number of relevant RA studies where randomization was performed but the comparator arm was not clearly distinguished in the registry or publication. All remaining procedures adhered to the original protocol and are documented in the OSF registry.

Results

Identification and inclusion of eligible studies

Our systematic search of Embase and MEDLINE yielded 960 records. After removing 14 duplicates, 946 unique titles and abstracts were screened using Rayyan. Of these, 670 were excluded for not meeting eligibility criteria. Full-text review was conducted for the remaining 276 articles, resulting in 28 additional exclusions based on study design, population, topic, date, or accessibility. A total of 248 studies underwent full data extraction and were included in the final analysis. Reasons for exclusion are detailed in figure 1.

Trial demographics and publication characteristics

The 248 included RCTs were published across a range of specialty journals. Anesthesia journals accounted for the largest proportion (44.0%; 109/248), followed by orthopedic surgery (26.6%; 66/248), general medicine journals (14.9%; 37/248), and pain, rehabilitation, or neurology journals (14.5%; 36/248). The median journal impact factor was 2.9 (IQR 1.6–4.4). Most trials were conducted in Asia (41.5%; 103/248) and Europe (26.2%; 65/248). The median sample size was 65 participants per trial (IQR 50–97). Funding was most frequently self-sourced (34.3%; 85/248), while 26.6% (66/248) did not include a funding statement. The highest number of included studies was published in 2023 (11.3%; 28/248). Additional journal characteristics are provided in table 2.

Table 2. General characteristics.

Characteristic	N=248
Impact factor of the journal, median (IQR)	2.9 (1.6–4.4)
Journal type, n (%)
Anesthesiology	109.0 (44.0)
Orthopedics	66.0 (26.6)
General medicine	37.0 (14.9)
Other*	36.0 (14.5)
Continent of trial conduct, n (%)
Asia	103.0 (41.5)
Europe	65.0 (26.2)
North America	43.0 (17.3)
Africa	15.0 (6.0)
Multiple	14.0 (5.6)
Australia	7.0 (2.8)
South America	1.0 (0.4)
Sample, median (IQR)	65.0 (50.0–97.0)
Funding source, n (%)
Self-funded/no outside funding	85.0 (34.3)
Funding not reported	66.0 (26.6)
Hospital	43.0 (17.3)
Multiple funding types	23.0 (9.3)
Government	19.0 (7.7)
Non-profit	6.0 (2.4)
Private/industry	4.0 (1.6)
University	2.0 (0.8)
Year, n (%)
2024	26.0 (10.5)
2023	28.0 (11.3)
2022	27.0 (10.9)
2021	23.0 (9.3)
2020	21.0 (8.5)
2019	24.0 (9.7)
2018	19.0 (7.7)
2017	17.0 (6.9)
2016	22.0 (8.9)
2015	17.0 (6.9)
2014	23.0 (9.3)
2013	1.0 (0.4)

^*The category ‘other’ consists of journals from the following disciplines: pain medicine, surgery, cardiovascular medicine, geriatrics, neurology, emergency medicine, environmental, healthcare sciences, multidisciplinary sciences, radiology, biochemistry, integrated and complementary medicine, oncology, otorhinolaryngology, pharmacy, and rheumatology.

Evaluation of clinical utility and transparency criteria

Each trial was assessed using a 13-item framework, which included 7 clinical utility criteria and 6 transparency criteria. Clinical utility and transparency values varied considerably across the 248 trials. Patient-centeredness was reported in all studies (100%; 248/248), and feasibility was met in 77.0% (191/248). Context placement (77.4%; 192/248) and information gain (71.0%; 176/248) were also common. In contrast, pragmatism (2.0%; 5/248) and value-for-money (0%; 0/248) were rare. For the problem-based criterion, 15.3% (38/248) of trials fully met the standard, 67.7% (168/248) partially met it, and 16.9% (42/248) did not meet it at all.

Transparency practices varied widely. While 88.7% (220/248) of trials disclosed conflicts of interest and 39.9% (99/248) were preregistered, only 10.1% (25/248) made their protocol publicly available. Public sharing of raw data was extremely limited (1.2%; 3/248). Additional details are presented in table 3.

Table 3. Criteria findings.

Clinical utility criteria	N=248	Transparency criteria	N=248
Problem base, n (%)		Pre-registration, n (%)
Full	38.0 (15.3)	Full	99.0 (39.9)
Partial	168.0 (67.7)	Partial	63.0 (25.4)
Absent	42.0 (16.9)	Absent	86.0 (34.7)
Context placement, n (%)		Public protocol, n (%)
Full	192.0 (77.4)	Full	25.0 (10.1)
Partial	52.0 (21.0)	Partial	3.0 (1.2)
Absent	4.0 (1.6)	Absent	220.0 (88.7)
Information gain, n (%)		Adherence to protocol, n (%)
Full	176.0 (71.0)	Full	28.0 (11.3)
Partial	0.0 (0.0)	Partial	1.0 (0.4)
Absent	72.0 (29.0)	Absent	219.0 (88.3)
Pragmatism, n (%)		Funding stated, n (%)
Full	5.0 (2.0)	Full	181.0 (73.0)
Partial	6.0 (2.4)	Partial	5.0 (2.0)
Absent	237.0 (95.6)	Absent	62.0 (25.0)
Patient centeredness, n (%)		Conflicts of interest, n (%)
Full	248.0 (100.0)	Full	220.0 (88.7)
Partial	0.0 (0.0)	Partial	0.0 (0.0)
Absent	0.0 (0.0)	Absent	28.0 (11.3)
Value for money, n (%)		Raw data, n (%)
Full	0.0 (0.0)	Full	3.0 (1.2)
Partial	0.0 (0.0)	Partial	0.0 (0.0)
Absent	248.0 (100.0)	Absent	245.0 (98.8)
Feasibility, n (%)
Full	191.0 (77.0)
Partial	1.0 (0.4)
Absent	56.0 (22.6)

Evaluation of context, pragmatism, and feasibility

Among the 248 included trials, 77.4% (192/248) referenced at least one prior systematic review to support their research context. A lack of feasibility was identified in 22.6% (56/248) of trials. Among these, most (17.7%; 44/248) did not offer any justification for feasibility concerns, while a smaller proportion cited specific barriers such as participant attrition (2.8%; 7/248) or slow recruitment (2.0%; 5/248). Pragmatism violations were widespread, occurring in 95.2% (236/248) of trials. The most frequent violations included single-center study design (88.7%; 220/248), assessor blinding (74.6%; 185/248), and presence of placebo control (18.1%; 45/248). Only 4.8% (12/248) of trials met all pragmatism criteria without any violations. Additional details related to systematic review inclusion, feasibility reporting, and pragmatism violations are presented in table 4.

Table 4. Systematic review inclusion, feasibility, and pragmatism in clinical trials.

Characteristic	N=248
Prior systematic review, n (%)
Systematic review is cited	192.0 (77.4)
Not cited or performed	56.0 (22.6)
Non-feasibility reason, n (%)
Study was feasible	192.0 (77.4)
No reason provided	44.0 (17.7)
Attrition	7.0 (2.8)
Low recruitment speed	5.0 (2.0)
Pragmatism violation*, n (%)
Single-centered trial	220.0 (88.7)
Blinding of assessors	185.0 (74.6)
Placebo-controlled	45.0 (18.1)
No pragmatism violation	12.0 (4.8)

^*One study had two pragmatism violations, resulting in a total N=462 for this section.

Changes in clinical utility and transparency over time

Clinical utility and transparency scores were evaluated across publication years from 2014 to 2024. As shown in figure 2, Panel A illustrates trends in clinical utility criteria, while panel B displays corresponding transparency values. These two domains were then combined to generate total usefulness scores, shown in panel C. While all three measures demonstrated a steady upward trajectory over time, the most substantial improvement was observed in transparency.

Figure 2. Trends in clinical utility and transparency values of RCTs on RA of lower extremity surgeries by year of publication. (A–C) The distribution of summed values for clinical utility criteria (A), transparency criteria (B), and total usefulness (C) across publication years. Each dot represents an individual RCT. Red dashed lines indicate fitted linear regression lines. RA, regional anesthesia; RCTs, randomized controlled trials.

Correlation between clinical utility and transparency

A statistically significant positive correlation was observed between clinical utility and transparency scores. As shown in figure 3, trials with higher transparency tended to have greater clinical utility. The Pearson correlation coefficient was r=0.32 (95% CI 0.20 to 0.42, t(246) = 5.25, p<0.001), indicating a low to moderate association.

Figure 3. Each point indicates a randomized controlled trial assessing regional anesthesia in lower extremity surgeries. The red dashed line represents a fitted linear regression. A positive, statistically significant correlation was observed between transparency and clinical utility values (Pearson’s r=0.32, 95% CI 0.20 to 0.42, t(246)=5.25, p<0.001).

Regression analysis of trial characteristics

Linear regression models were used to identify predictors of clinical utility, transparency, and total usefulness scores. As shown in table 5, both journal impact factor (β=0.25, p=0.01) and self-funding or absence of outside funding (β=0.93, p=0.02) were significantly associated with higher clinical utility scores. No other predictors,including sample size, continent of trial conduct, or journal type, were statistically significant. Bolded values are statistically important.

Table 5. Linear regression: usefulness value.

Predictor	Estimate	SE	Statistic	P value	95% CI
(Intercept)	6.84	0.55	12.50	0.00	(5.76 to 7.93)
Sample	<0.01	<0.01	−1.00	0.32	(−0.01 to 0.00)
Continent of trial conduct
Europe	−0.46	0.36	−1.26	0.21	(−1.18 to 0.26)
North America	0.01	0.43	0.03	0.98	(−0.83 to 0.85)
Journal type
General medicine	−0.45	0.48	−0.93	0.35	(−1.41 to 0.51)
Orthopedics	0.07	0.37	0.19	0.85	(−0.65 to 0.80)
Funding source
Hospital	0.66	0.43	1.54	0.13	(−0.19 to 1.50)
Self-funded/no outside funding	0.93	0.39	2.42	0.02	(0.17 to 1.69)
Impact factor of the journal	0.25	0.09	2.77	0.01	(0.07 to 0.43)

For transparency (table 6), several predictors emerged. Self-funded or unfunded trials were most strongly associated with improved transparency (β=3.18, p<0.01), followed by hospital-funded trials (β=2.57, p<0.01). Trials conducted in North America (β=1.08, p=0.02) and those published in higher-impact journals (β=0.25, p=0.01) also demonstrated significantly greater transparency. Journal type and sample size were not significant predictors.

Table 6. Linear regression: transparency value.

Predictor	Estimate	SE	Statistic	P value	95% CI
(Intercept)	1.42	0.58	2.46	0.02	(0.28 to 2.56)
Sample	<0.01	<0.01	1.21	0.23	(−0.00 to 0.01)
Continent of trial conduct
Europe	0.57	0.38	1.48	0.14	(−0.19 to 1.33)
North America	1.08	0.45	2.41	0.02	(0.19 to 1.97)
Journal type
General medicine	0.17	0.51	0.33	0.74	(−0.84 to 1.18)
Orthopedics	−0.47	0.39	−1.22	0.22	(−1.24 to 0.29)
Funding source
Hospital	2.57	0.45	5.74	<0.01	(1.69 to 3.46)
Self-funded/no outside funding	3.18	0.41	7.82	<0.01	(2.37 to 3.98)
Impact factor of the journal	0.25	0.10	2.64	0.01	(0.06 to 0.44)

In the model for total usefulness (table 7), all three of these factors remained significant. Self-funded trials (β=4.11, p<0.01), hospital-funded trials (β=3.23, p<0.01), and higher journal impact factor (β=0.51, p<0.01) were all independently associated with greater overall usefulness

Table 7. Linear regression: total usefulness value.

Predictor	Estimate	SE	Statistic	P value	95% CI
(Intercept)	8.26	0.83	9.90	<0.01	(6.61 to 9.91)
Sample	<0.01	<0.01	0.19	0.85	(−0.01 to 0.01)
Continent of trial conduct
Europe	0.11	0.55	0.20	0.84	(−0.99 to 1.21)
North America	1.09	0.65	1.68	0.10	(−0.20 to 2.37)
Journal type
General medicine	−0.28	0.74	−0.38	0.70	(−1.74 to 1.18)
Orthopedics	−0.40	0.56	−0.72	0.47	(−1.51 to 0.70)
Funding source
Hospital	3.23	0.65	4.98	<0.01	(1.94 to 4.52)
Self-funded/no outside funding	4.11	0.59	6.99	<0.01	(2.94 to 5.27)
Impact factor of the journal	0.51	0.14	3.64	<0.01	(0.23 to 0.78)

Discussion

Major findings

This systematic review of 248 RCTs on RA for lower-extremity surgery shows steady gains in transparency but persistent shortcomings in clinical utility. Although total usefulness scores trended upward, major gaps remain, particularly in pragmatism, value-for-money analysis, and open data practices, which limit real-world adoption of RA evidence. Compared with a recent Regional Anesthesia and Pain Medicine analysis of spinal cord stimulation RCTs that found frequent undisclosed conflict of interests, our review observed fewer omissions, suggesting incremental improvement in RA trial reporting.²²

Several factors may underlie these persistent gaps. Academic incentives often reward smaller, tightly controlled trials that use surrogate outcomes and are easier to publish, while pragmatic or economic evaluations are more complex and resource-intensive. Limited funding also restricts investigators’ ability to conduct multi-center studies, incorporate cost analyses, or extend follow-up. Practical barriers such as patient retention, regulatory requirements, and lack of infrastructure further discourage long-term outcome assessment and data sharing.

Implications for clinical practice

Encouragingly, all trials addressed patient-centered outcomes, and most met feasibility and context placement standards. However, restrictive eligibility criteria, often excluding patients with obesity, neuropathy, or opioid use, limit external validity. Short follow-up, typically 24–48 hours, also fails to capture complications such as rebound pain or catheter-related issues.^6–8

These findings echo longstanding patterns identified in meta-research across multiple clinical domains where trials often fall short on pragmatic design, generalizability, economic evaluation, and inclusive recruitment strategies.²³ In RA, the International Consensus on Anesthesia-Related Outcomes after Surgery (ICAROS) group has called for trials that reflect the complexity of contemporary surgical populations by enrolling patients with diverse comorbidities, varying baseline functional status, and different care contexts.²⁴ Their recommendations go beyond participant selection to include trial features such as standardized reporting of core perioperative outcomes, longer follow-up periods, and alignment with rehabilitation protocols that more accurately reflect how anesthetic choices influence recovery trajectories. Yet, most RA trials remain confined to academic centers, with narrow inclusion criteria and limited external relevance.

Implications for research and policy

Organizations like ICAROS and similar consortia are uniquely positioned to catalyze a shift toward more useful trial designs. As bridges between clinicians, methodologists, funders, and patients, they can support core development, encourage use of tools like PRECIS-2 to assess pragmatism, and push for funder alignment on value-of-information (VOI) modeling and long-term follow-up.²⁵ They can also serve as repositories for best practices, protocols, and anonymized datasets, promoting harmonization and reducing redundancy. This collective infrastructure lowers the burden of conducting high-quality trials, especially for underfunded or community-based teams lacking the resources of large academic centers.

To strengthen the real-world impact of RA research, future trials must reconsider both who is enrolled and how success is measured. Patient-centered endpoints such as time to ambulation, discharge readiness, opioid-free days, and functional independence should replace traditional surrogates like block onset time.¹⁰ Collaboration among anesthesiologists, physical therapists, and surgeons can support more holistic trial designs that extend beyond the post-anesthesia care unit.²⁶ Embedding these studies in routine workflows through registry-based or platform models offers a practical path forward by reducing cost, streamlining recruitment, and ensuring results are applicable across diverse care settings.

Strengths, limitations, and future directions

Despite rising demand for evidence-based RA, none of the 248 trials incorporated VOI or budget impact analyses. VOI modeling can help funders determine whether new block techniques merit investment or whether existing evidence suffices, while budget impact models estimate downstream costs and savings such as reduced opioid use or shorter hospital stays.²⁷ Without these tools, RA research risks misalignment with healthcare system priorities. Incorporating them would align trials with health technology assessment standards and ensure findings resonate with clinicians, hospitals, and policymakers.^{28 29}

Our regression analysis clarified study characteristics linked with higher usefulness scores and highlighted directions for future trial design. Trials published in higher-impact journals consistently showed greater clinical utility and transparency, suggesting that editorial standards may promote stronger methodology and reporting.³⁰ Moreover, funding source emerged as a significant predictor: hospital-funded and self-funded trials outperformed other funding models, including government and industry, in overall usefulness. This may reflect the alignment of hospital-supported research with practical clinical needs and the autonomy of self-funded teams to pursue open science principles without commercial pressures.³¹ Geography mattered as well, with North American trials demonstrating higher transparency, possibly due to National Institutes of Health policies, Food and Drug Administration requirements, and regional norms around pre-registration and data sharing.³²

Our review has several strengths, including pre-registration, duplicate masked extraction, and the public availability of all data and code. The van ’t Hooft framework enabled an assessment of trial quality, capturing both methodological robustness and reporting transparency. Limitations should be acknowledged. Usefulness scoring inherently involves subjective judgments, and although adjudication processes were in place, misclassification is possible. A risk of bias assessment was not conducted, which can limit transparency for policy decision-making. Restriction to English-language trials may limit generalizability, and the overrepresentation of single-center Asian trials could skew estimates. Furthermore, while the van ’t Hooft framework has conceptual appeal, it has not been formally validated for use in RA-specific research. Finally, our assessment of conflict-of-interest disclosure relied solely on what was explicitly reported in trial manuscripts; external registries (eg, Open Payments) were not queried, and undisclosed conflicts may therefore have gone unrecognized.

Conclusion

Despite gains in transparency, RCTs on RA for lower-extremity surgery continue to underdeliver on elements essential for clinical adoption, especially pragmatism, data sharing, and value-for-money analysis. To narrow the gap between evidence generation and use, stakeholders must align around open science mandates, pragmatic trial design, and the development of standardized, patient-relevant outcomes. This will elevate the quality of RA research and ensure findings meaningfully inform anesthetic care for diverse surgical populations.