High Variability in Arthroscopic Fluid Use Among Surgeons and Joints for Common Orthopaedic Procedures

Abstract

Purpose

To determine arthroscopic fluid volume use patterns across common elective orthopaedic procedures, evaluate intersurgeon consistency in fluid use, and assess whether blood loss reduction adjuncts affect fluid consumption to generate actionable insights to guide surgeons in optimizing resource use, reducing waste, and improving efficiency during times of resource constraint.

Methods

A cross-sectional analysis at a single outpatient surgery center of adult arthroscopic procedures was conducted. A representative sample of 500 arthroscopic cases of the shoulder, elbow, knee, and ankle performed by multiple surgeons were reviewed. A total of 449 cases met inclusion criteria. Analysis of fluid use relative to body area, Current Procedural Terminology codes, surgeon, and blood loss reduction adjuncts (tranexamic acid [TXA] and epinephrine) was performed using analysis of variance tests.

Results

Mean fluid volume was greatest in shoulder arthroscopy (16,582 ± 7,026 mL), followed by knee (8,677 ± 7,027 mL), elbow (6,489 ± 2,540 mL), and ankle procedures (3,534 ± 1,500 mL). Significant intersurgeon variation was observed (F = 19.61, P = .0001). Epinephrine was used in 58.8% of cases, with mean fluid volume of 11,571 ± 6,896 mL for local injection and 15,401 ± 8,610 mL for arthroscopic fluid administration. Cases without epinephrine averaged 7,676 ± 6,570 mL of fluid use. TXA was used in only 0.7% of cases. Neither epinephrine nor TXA showed significant reduction in fluid consumption (P > .05).

Conclusions

Arthroscopic fluid use across anatomic regions was highly varied, with shoulder procedures requiring the greatest volumes. Surgeon-specific use patterns also varied considerably. The use of neither TXA nor epinephrine reduced fluid consumption, although this finding may reflect the small sample size and lack of standardized protocols rather than true ineffectiveness.

Level of Evidence

Level III, cross-sectional study.

The national shortage of intravenous fluids in the United States that began in 2024 has emerged as a critical issue and has impacted patient care at hospitals and ambulatory surgery centers. The fluid shortage has resulted in fluid rationing, procedure cancellations, and delays in care.^1,2 Attempts to optimize pump pressure and portal placement to facilitate efficient irrigation have previously been published.^3,4 However, data on expected fluid use for standard arthroscopic orthopaedic procedures are less well known, leaving decision makers without a reliable benchmark of the fluid “cost” for each procedure. As supply chain issues and supply shortages have become more commonplace,^5,6 it is important for hospital systems and ambulatory surgery centers alike to have more evidence-based approaches to resource acquisition and planning.

Furthermore, multiple techniques have been described to limit blood loss and improve visualization during arthroscopy—such as the use of tranexamic acid (TXA) and epinephrine.7, 8, 9, 10 Although these methods have been shown to reduce total operative time in certain procedures,^7,11,12 it is less clear to what extent these interventions influence the total volume of arthroscopy fluid used. Bermudo Gamboa et al.¹¹ found no significant reduction in the volume of fluid used during shoulder arthroscopy with the use of preoperative local epinephrine infiltration but did note improved visualization. Similarly, Jensen et al.¹³ found improved visualization but no difference the amount of irrigation fluid when diluting epinephrine in saline via a pressure-controlled pump used intraoperatively.

Anecdotally, it is common practice to use 3-L bags of fluid during arthroscopy. However, there is very little information regarding the amount of fluid used in arthroscopic cases other than isolated studies such as one by Smith and Shah¹⁴ in 2008, which showed the mean fluid use of 3.2 ± 2.2 L for shoulder arthroscopy. However, this may not be the most efficient size to use during a national shortage of fluid. In addition, although data exist regarding the overall cost of arthroscopic procedures across different joints, these studies often focus on broader comparisons, such as cost differences between surgery centers and hospital-based practices or variations by Current Procedural Terminology (CPT) codes or implant choices, seldom diving into fluid use and its costs.14, 15, 16, 17, 18, 19 The purpose of this study is to determine arthroscopic fluid volume use patterns across common elective orthopaedic procedures, evaluate intersurgeon consistency in fluid use, and assess whether blood loss reduction adjuncts affect fluid consumption to generate actionable insights to guide surgeons in optimizing resource use, reducing waste, and improving efficiency during times of resource constraint. We hypothesized that (1) arthroscopic fluid use would vary significantly by anatomic region for common elective orthopaedic procedures, (2) different surgeons would have consistent fluid use patterns when performing similar procedures, and (3) the use of blood loss reduction adjuncts (TXA and epinephrine) would be associated with reduced arthroscopic fluid consumption.

Methods

Study Design and Setting

This retrospective cross-sectional study was conducted at a single outpatient surgery center that is owned and used by a large multispecialty private practice orthopaedic surgery group. The group of surgeons includes a mix of general orthopaedic surgeons and fellowship-trained subspecialists typical of a community orthopaedic practice. Elective arthroscopic surgeries performed between 2014 and 2024 were compiled for screening. Data were collected retrospectively from a review of the electronic medical record. Data were entered securely in an institution-specific REDCap database. This study received approval from local institutional review board (IRB00121568 approved on November 7, 2024, by Wake Forest University Institutional Review Board) and is compliant with US Health Insurance Portability and Accountability Act (HIPAA).

Participants

The study included adult patients (≥18 years) undergoing arthroscopic orthopaedic surgeries between 2014 and 2024. Patients were excluded if there was no documentation of total volume of arthroscopy fluid use. All records included for initial review were sorted into random sequence before being uploaded to the REDCap database, the first 500 were then selected as a representative random sample—with subsets allocated by body area (200 shoulders, 200 knees, 50 elbows, and 50 ankles) (Fig 1). The number of cases for each body area was predetermined on the basis of the distribution of surgical volume of the surgical center by body area, the inherent variability of CPT codes used in each body area (greater for shoulder and knee than ankle and elbow), and what could be pragmatically completed in a timely manner while the national fluid shortage was still a current issue.

Fig 1.

CONSORT diagram of patient selection for arthroscopic fluid use analysis. From 3,937 total surgeries assessed for eligibility, 3,434 were randomly excluded to create a representative sample of 501 surgeries (40% knee, 40% shoulder, 10% elbow, 10% ankle). After excluding 1 nonarthroscopic surgery and 51 cases with missing data (21 without documented arthroscopy fluid use, 27 patients younger than 18 years, 3 with multiple exclusion criteria), 449 eligible surgeries were included in the final analysis (200 shoulder, 159 knee, 46 elbow, 44 ankle). This systematic selection process ensured representative sampling across all major arthroscopic joint procedures while maintaining adequate statistical power for intersurgeon and interjoint comparisons. (CONSORT, Consolidated Standards of Reporting Trials.)

Description of Experiment

After screening for inclusion, data pertinent to the arthroscopic fluid volume were collected via retrospective review of the electronic medical record, including surgeon, CPT codes, use of epinephrine and TXA, volume of arthroscopic fluid used, and arthroscopic pump fluid pressure when available.

Variables, Outcome Measures, Data Sources, and Bias

The primary outcome was the volume of arthroscopic fluid used per procedure. Independent variables included surgeon, CPT codes, patient positioning, and the use of TXA or epinephrine. Epinephrine was reported as administered via local intra-articular injection and/or diluted in arthroscopic fluid and administered intraoperatively via pressure-controlled pump. TXA was given only intravenously. Potential biases, such as inconsistencies in data recording, were minimized through standardized data-abstraction protocols.

Statistical Analysis

Descriptive statistics of means standard deviations, minimums and maximums, and frequencies were used to summarize the data. F ratio statistics from analysis of variance tests were used to determine significant differences between the means. Study data were collected and managed using Research Electronic Data Capture tools hosted locally.^20,21 All statistical analyses were conducted using SAS/STAT software, Version 9.4, of the SAS System for Windows (SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc., Cary, NC). A power calculation was not performed because of the retrospective and cross-sectional nature of the study.

Accounting for All Patients/Study Subjects

All eligible patients during the study period were included, with bilateral cases analyzed separately when applicable to ensure comprehensive data representation.

Results

A total of 449 surgeries in 444 patients met final inclusion criteria. Reasons for exclusion included no arthroscopy used (n = 1), no documented use of fluid (n = 21), patients younger than 18 years old (n = 27), or more than 1 of the aforementioned causes (n = 3). Of the final study population, the mean age at the time of surgery was 51 ± 15 years, and most patients were male (n = 271, 61%). The most common operative site was the shoulder (n = 200, 44.5%) followed by the knee (n = 159, 35.4%) (Table 1).

Table 1.

Demographics

Characteristics	N = 444 (%)
Sex
Female	173 (38.9)
Male	271 (61.1)
Race
White	318 (71.6)
Black	67 (15.1)
Unknown or declined	52 (11.7)
Hispanic	6 (1.4)
Asian or Native American	1 (0.2)
Age at surgery, yr, mean (SD)	51 (14.9)
BMI, mean (SD)	29.8 (5.9)

BMI, body mass index; SD, standard deviation.

Arthroscopic Fluid Use by Anatomic Region and CPT Code

The mean volume of arthroscopic fluid used for all included surgeries was 11,442 mL. The mean volume of arthroscopic fluid used was greatest for shoulder surgeries (16,582 mL) and lowest for ankle surgeries (3,534 mL) (Table 2). The difference in mean arthroscopic fluid use by anatomic region was statistically significant (P < .001), with varying effect sizes of individual CPT codes on overall fluid use (Table 3). There was insufficient variety of CPT codes used in ankle arthroscopy to calculate an effect size with only a single CPT code being used (29898).

Table 2.

Fluid Used by Anatomic Region

Anatomic Region	N = 449	Mean (SD)	Range
Shoulder	200	16,582 (7,026)	3,000-51,000
Elbow	46	6,489 (2,540)	3,000-12,000
Knee	159	8,677 (7,027)	3,000-39,000
Ankle	44	3,534 (1,500)	2,000-9,000
Total	449	11,442 (8,538)	2,000-51,000

SD, standard deviation.

Table 3.

Effect on Total Fluid Volume of Individual CPT Codes

CPT Code	CPT Description	N	Effect on Mean	P Value
Shoulder arthroscopy (mean volume: 16,582 mL)
29806	Arthroscopy, shoulder, surgical; capsulorrhaphy	2	+ 4,977 mL	.47
29825	Arthroscopy, shoulder, surgical; with lysis and resection of adhesions, with or without manipulation	3	− 4,646 mL	.34
29826	Arthroscopy, shoulder, surgical; decompression of subacromial space with partial acromioplasty, with coracoacromial (ie, arch) release, when performed	141	− 533 mL	.73
29827	Arthroscopy, shoulder, surgical; with rotator cuff repair	195	+ 3,490 mL	.61
29828	Arthroscopy, shoulder, surgical; biceps tenodesis	130	+ 3,411 mL	.02
29806	Arthroscopy, shoulder, surgical; capsulorrhaphy	2	+ 4,977 mL	.47
29825	Arthroscopy, shoulder, surgical; with lysis and resection of adhesions, with or without manipulation	3	− 4,646 mL	.34
29826	Arthroscopy, shoulder, surgical; decompression of subacromial space with partial acromioplasty, with coracoacromial (ie, arch) release, when performed	141	− 533 mL	.73
29827	Arthroscopy, shoulder, surgical; with rotator cuff repair	195	+ 3,490 mL	.61
Elbow arthroscopy (mean volume: 6,489 mL)
29834	Arthroscopy, elbow, surgical; with removal of loose body or foreign body	16	+ 1.301 mL	.13
29835	Arthroscopy, elbow, surgical; synovectomy, partial	1	+ 4,858 mL	.04
29836	Arthroscopy, elbow, surgical; synovectomy, complete	3	+ 3,292 mL	.02
29837	Arthroscopy, elbow, surgical; debridement, limited	1	− 3,751 mL	.14
29838	Arthroscopy, elbow, surgical; debridement, extensive	41	+ 391 mL	.77
Knee arthroscopy (mean volume: 8,677 mL)
29874	Arthroscopy, knee, surgical; for removal of loose body or foreign body (eg, osteochondritis dissecans fragmentation, chondral fragmentation)	5	− 1,067 mL	<.01
29877	Arthroscopy, knee, surgical; debridement/shaving of articular cartilage (chondroplasty)	4	+ 635 mL	.73
29880	Arthroscopy, knee, surgical; with meniscectomy (medial AND lateral, including any meniscal shaving) including debridement/shaving of articular cartilage (chondroplasty), same or separate compartment(s), when performed	13	+ 4,876 mL	.86
29882	Arthroscopy, knee, surgical; with meniscus repair (medial OR lateral)	58	+ 1,384 mL	.24
29883	Arthroscopy, knee, surgical; with meniscus repair (medial AND lateral)	3	+ 9,001 mL	.03

Arthroscopic Fluid Variability by Surgeons

There was wide variability across surgeons—depending on the body region (Table 4). The mean consumption for different surgeons in shoulder arthroscopy ranged from 10,429 to 17,001 mL (Fig 2). In the knee, it ranged from 4,800 to 11,500 mL (Fig 3). Elbow means between surgeons ranged from 5,346 to 12,000 mL. Ankle arthroscopies were only performed by 2 surgeons, with means ranging from 3,000 to 3,547 mL. The difference in mean arthroscopic fluid use by surgeon was statistically significant (P < .001).

Table 4.

Fluid Use by Surgeon

Surgeon	N = 449	Mean (SD)	Range
Surgeon A
Shoulder	1	15,000 (−)	−
Elbow	1	12,000 (−)	−
Knee	18	5,500 (2,956)	3,000-15,000
Surgeon B
Shoulder	66	14,890 (5,103)	3,000-24,004
Elbow	16	6,177 (1,976)	3,000-9,003
Surgeon C
Shoulder	5	13,200 (7,530)	6,000-21,000
Elbow	1	3,000 (−)	−
Knee	94	7,807 (6,002)	3,000-30,000
Surgeon D
Shoulder	9	17,001 (4,500)	12,000-27,000
Elbow	12	7,750 (2,701)	3,000-12,000
Surgeon E
Knee	5	6,000 (2,121)	3,000-9,000
Surgeon F
Shoulder	25	12,121 (7817)	3,000-45,000
Elbow	2	8,002 (1,416)	7,000-9,003
Knee	1	6,000 (−)	−
Surgeon G
Knee	12	11,500 (1,0917)	3,000-39,000
Shoulder	26	21,346 (7,216)	12,000-36,000
Surgeon H
Knee	8	7,875 (4,518)	3,000-18,000
Shoulder	40	20,863 (10,535)	3,000-48,000
Surgeon I
Shoulder	21	10,429 (4,894)	3,000-21,000
Elbow	13	5,346 (2,249)	3,000-9,500
Surgeon J
Knee	5	4,800 (1,643)	3,000-6,000
Surgeon L
Ankle	43	3,547 (1,515)	2,000-9,000
Other surgeons (n < 5)
Knee	18	5,500 (2,956)	3,000-15,000
Shoulder	1	15,000 (−)	−
Elbow	1	12,000 (−)	−
Ankle	1	3,000 (−)	−

SD, standard deviation.

Fig 2.

Violin plots showing substantial intersurgeon variability in arthroscopic fluid consumption (milliliters) during shoulder procedures across 9 surgeons (A-I). Each violin represents the distribution of fluid use for an individual surgeon, with width indicating frequency of cases at each volume level. Surgeon A performed only 1 case (15,000 mL), while surgeons with larger case volumes (B, C, G, H, I) showed distinct use patterns ranging from conservative (Surgeon I: mean 10,429 mL) to high-volume (Surgeon G: mean 21,346 mL). The substantial variation (>2-fold difference between highest and lowest mean use) suggests significant practice pattern differences that may represent opportunities for standardization and resource optimization during fluid shortages.

Fig 3.

Violin plots showing inter-surgeon variability in arthroscopic fluid consumption (milliliters) during knee procedures across 7 surgeons (A, C, E, G, H, J, Other). Each violin shows the distribution of fluid use per surgeon, with violin width reflecting case frequency at each volume. Surgeon C performed the most knee cases (94) with moderate use (mean 7,807 mL), while Surgeon G showed the highest mean consumption (11,500 mL) and Surgeon J the lowest (4,800 mL). The >2-fold variation between surgeons performing similar knee arthroscopic procedures indicates substantial practice pattern differences that could be targeted for standardization protocols to improve resource use efficiency.

Blood Loss Adjuncts

Epinephrine either via local injection or in arthroscopic fluid was used as a surgical adjunct in 320 (71.3%) of the included surgical cases. In 56 cases (12.5%), epinephrine was given both via local injection and in arthroscopic fluid. The mean arthroscopic fluid used when epinephrine was given via local injection and in arthroscopic fluid was 11,571 ± 6896 mL and 15,401 ± 8,610 mL, respectively. The mean volume of fluid used in cases without any epinephrine was 7,676 ± 6570 mL, which was significantly lower compared with epinephrine given in arthroscopic fluid (P < .001). However, no significant difference in mean fluid volume was found between cases without epinephrine and cases with epinephrine given via injection (P = 0.671). Intraoperative use of TXA occurred in 3 (0.7%) cases (Table 5).

Table 5.

Blood Loss Adjuncts

Adjunct	n	Mean (SD)	Range
Tranexamic acid
Yes	3	26,000 (13,528)	12,000-39,000
No	441	11,381 (8,433)	2,000-51,0000
Missing	5	−	−
Epinephrine
In arthroscopic fluid	221	15,401 (8,610)	3,000-5,1000
Through injection	99	11,571 (6,896)	3,000-30,000
None	129	7,676 (6,570)	3,000-24,004

SD, standard deviation.

Discussion

Our findings reveal distinct variations and patterns in fluid use across different anatomic regions and common orthopaedic procedures. Shoulder arthroscopy consistently required the greatest volume (16,582 mL), using nearly twice the fluid of knee procedures (8,677 mL) and substantially exceeding the previously reported mean of 3.2 ± 2.2 L.¹⁴ This marked difference likely reflects the variety of our case mix and the inclusion of various arthroscopic procedures, as evidenced by the effect of several CPT codes on fluid volume. Furthermore, shoulder arthroscopy showed the widest variability across procedures and surgeons. This case-by-case variability in shoulder procedures is consistent with a previous meta-analysis which outlined variability of fluid use during shoulder arthroscopy often attributed to differences in applied pump pressure and degree of extravasation that can occur.²² The primary anatomical factors for this are believed to be looser subcutaneous soft tissues and a lack of encapsulation in the subacromial space, which allows fluid to more readily disperse into surrounding tissues.²³ In contrast, ankle arthroscopy showed both low total fluid use and low variability. This should be interpreted with some caution, however, given the limited number of surgeons who performed ankle arthroscopy in this study, with a single CPT code used. These findings have important implications for resource management and cost efficiency of arthroscopic procedures. The current practice of using standard 3-L bags appears inefficient, given the wide variation in procedure-specific requirements, potentially contributing to unnecessary waste. Previous studies have estimated that operating rooms alone can constitute up to 70% of hospital-based waste, with arthroscopic procedures being a major contributor.^24,25 Our data suggest an opportunity to stratify fluid allocation by procedure type or duration, implementing procedure-specific fluid packs that could significantly reduce waste. This approach could be particularly impactful for lower-volume, lower-variability procedures such as ankle arthroscopy, where mean use (3,534 mL) suggests that 1-L bags may suffice instead of traditional 3-L bags. However, there are potential unintended consequences of transitioning to 1-L bags, such as increased task burden on operating room ancillary staff and risk of lapses in fluid during arthroscopic cases. Such targeted allocation strategies could reduce the economic and environmental burden of fluid waste in hospitals and surgery centers. The potential significance of optimizing fluid allocation could be substantial when considering the totality of arthroscopic procedures performed annually in the United States.

The marked variation in fluid use between surgeons performing similar procedures represents a particularly striking finding. With a greater than 5-fold difference between the highest and lowest volume users (20,348 mL vs 3,547 mL), our data suggest that individual surgical specialties and case makeup significantly influence resource use. This variation was attenuated but still present when controlling for anatomic region. Variation of still greater than a 2-fold difference persisted when considering arthroscopy of the same body region between providers. Although some variation may be justified by case complexity, intraoperative findings, or patient factors, the magnitude of difference observed suggests other factors may be at play—such as arthroscopic pump pressure, which was not recorded in patients in this study. This aligns with broader literature reporting significant intersurgeon variability physician practice patterns including arthroscopic cases.26, 27, 28 Designed clinical pathways, regular feedback to surgeons about their use patterns, and peer-to-peer education programs could be several options targeted towards reducing unwarranted variation while respecting necessary procedural differences and complexities. Previous successful standardization initiatives in orthopedic surgery suggest that specific protocols, clinical pathways, regular performance feedback and peer-to-peer education can effectively reduce unwarranted practice variation without compromising surgical autonomy or outcomes.^29,30 During the fluid shortage, our institution made several recommendations to surgeons to minimize fluid consumption during arthroscopic procedures: (1) controlled hypotensive anesthesia; (2) consistent pump pressures at normal to low range; (3) use of “day tubing”; (4) liberal use of cannulas; (5) use of 1-L fluid bags. Further study is needed to determine if these modifications ultimately led to less fluid consumption. Of these recommendations, perhaps the most significant change was the use of “day tubing” which allows for use of fluid from one patient to the next on a single day eliminating the waste of unused fluid especially with 3-L bags.

Despite literature reporting reduced operative times with adjunct use,^11,12 adjuncts did not correlate with reduced fluid requirements in our cohort. This aligns with a previous study, which reported improved visualization without notable reductions in fluid consumption.¹³The use of epinephrine in arthroscopic fluid had an overall greater total volume of fluid used compared with cases not using epinephrine in any form. This paradoxical increase may be due to a larger proportion of high-volume cases selectively using epinephrine, such as shoulder arthroscopy, which had significantly greater volumes of total fluid used when compared with other anatomic region (knee, ankle, elbow). In addition, the minimal adoption of TXA (0.7%) in our cohort contrasts with growing evidence supporting its efficacy.8, 9, 10 Although epinephrine was more commonly administered (58.8%), neither local injection nor arthroscopic fluid addition showed superior fluid conservation. These findings are of limited conclusive value and more robust studies should be complete to determine the causal impact of these adjuncts on overall fluid consumption.

Limitations

Several limitations of this study warrant discussion. First, as a retrospective analysis conducted at a single outpatient surgery center, our findings may not be generalizable to all practice settings or patient populations. In addition, our ability to standardize and control confounding variables such as case complexity and patient-specific factors was limited, which prevents us from drawing causal inferences and rather serves as a reported observation. The absence of pump pressure data represents a significant limitation as pressure settings directly influence flow rates and total fluid consumption. Furthermore, the absence of data on operative time for the included cases precludes the ability to investigate the association between surgical time and arthroscopic fluid consumption. Conclusions from our analysis of hemostatic agents are limited by several factors. The small sample size and no standardized protocol for use of hemostatic agents allows for substantial bias and these findings should be interpreted with caution.

Conclusions

Arthroscopic fluid use across anatomic regions was highly varied, with shoulder procedures requiring the greatest volumes. Surgeon-specific use patterns also varied considerably. Neither TXA nor epinephrine reduced fluid consumption, although this may reflect the small sample size and lack of standardized protocols rather than true ineffectiveness.

Disclosures

All authors (A.B.R., C.C.C., M.D., E.M.P., A.B.W., K.K.G., N.H.) declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.