Can Robots Keep You Upright? An Ergonomic Analysis of Surgeon Posture in Robotic Versus Conventional Total Knee Arthroplasty

Abstract

Background: Robotic-assisted technology has become an increasingly utilized adjunct within the realm of primary total knee arthroplasty (TKA). Previous studies have shown that robotic-assisted total knee arthroplasty (raTKA) offers potential advantages of enhanced bony preparation and optimal implant alignment with equivalent long-term patient outcomes and component longevity in comparison to conventional TKA (cTKA). Furthermore, recent studies have identified the additional benefit of decreased surgeon physiologic stress with the use of raTKA. The purpose of this study was to compare differences in surgeon posture between primary raTKA and cTKA. Materials and Methods: We prospectively evaluated 103 consecutive primary TKA cases (48 raTKAs, 55 cTKAs) performed by three high-volume, fellowship-trained arthroplasty surgeons. Throughout each case, surgeons wore a posture-tracking device to evaluate time spent slouching. The threshold for slouching was set to 30 degrees of flexion from a neutral spinal axis. Demographic and operative factors were collected. Two-tailed tests and multivariate analysis were used to assess for differences between groups. Results: After controlling for individual surgeon differences in posture, we found a decrease in the percentage and duration of time spent slouching in raTKA cases compared to cTKA cases (42.4 vs. 72.5%, p < 0.001, 35.4 vs. 54.7 min, p = 0.037). On average, the use of the robot decreased surgeon slouching time by 19.3 min (26.6%, p < 0.001). Patient factors such as increased age and ASA 2 were also associated with favorable effects on posture (p < 0.001). Conclusions: Surgeons performing primary raTKA cases spend significantly less case time and case percentage in a slouched posture compared to conventional primary TKA cases. This suggests the potential for ergonomic benefit of robotic-assisted technology in primary TKA. Further research is needed to determine the long-term effects of posture on surgeon pain and career longevity.

1. Introduction

Work-related musculoskeletal symptoms (WRMS) represent a growing and underestimated problem in the surgical field [1]. Ergonomists have described WRMS in surgeons to be similar, if not greater, than those exhibited by certain industrial workers, including coal miners and manufacturing laborers [2]. Orthopedic surgery is no exception. It is a physically demanding field that places stress on the musculoskeletal system from repetitive motions, prolonged standing, and operating in non-ergonomic positions [3,4,5,6,7,8]. There is a high incidence of back and neck pain caused by static head and back-bent postures [9,10]. Auerbach et al. reported a career prevalence of 27.6% for cervical radiculopathy among orthopedic surgeons, of whom 10.7% required surgical intervention and 18.9% required time off work, ranging from days to forced early retirement [11].

Among the subspecialties of orthopedics, adult reconstruction may be the most laborious [6,7,9,10]. In a survey of 586 arthroplasty surgeons, McQivey et al. found that 96.5% experienced procedure-related musculoskeletal pain, with lower back and neck comprising 34.2% and 21.2% of complaints, respectively [10]. In another series, Alqahtani et al. found that over a quarter of arthroplasty surgeons surveyed required time off work at one point due to a work-related injury and that over 60% have reported a work-related injury during their career [6].

Robotic-assisted technology has become an increasingly utilized adjunct within the realm of primary total knee arthroplasty (TKA) [12,13]. Previous studies have shown that robotic-assisted total knee arthroplasty (raTKA) offers potential advantages of enhanced bony preparation and optimal implant alignment with equivalent long-term patient outcomes and component longevity in comparison to cTKA (cTKA) [14,15]. Moreover, a recent study by Haffar et al. suggested that raTKA results in less surgeon physiological stress and energy expenditure compared to cTKAs. This study also demonstrated some postural benefit to raTKA, as the robot decreased surgeon mean lumbar flexion time significantly [16]. These findings align with research in other medical fields, such as gynecology and urology, where robotic assistance has been shown to improve surgeon posture and reduce musculoskeletal strain.

Given the substantial burden of WRMS among orthopedic surgeons, further exploration into ergonomic interventions is warranted. The purpose of this study is to compare differences in surgeon posture while performing primary raTKA and cTKA, as well as to identify additional patient or operating room factors that may influence surgeon ergonomics. By analyzing a larger cohort of surgeons, we aim to provide deeper insights into strategies that could enhance workplace ergonomics, improve surgeon well-being, and ultimately extend career longevity in orthopedic surgery.

2. Materials and Methods

2.1. Data Collection

Following Institutional Review Board approval, we prospectively evaluated postural data on three high-volume, fellowship-trained arthroplasty surgeons while performing 103 consecutive primary TKA cases (48 raTKAs, 55 cTKAs). The three surgeons were at varying stages of their careers, ranging from 3 to over 20 years in practice. Perioperative postural data were collected for all cases via a posture tracking device (UPRIGHT GO, Tel Aviv, Israel) worn by the surgeon throughout the case. At the start of each case, the operative room assistant began device data collection via the mobile device at the time of incision and terminated data collection for the case immediately prior to closing the arthrotomy. The wearable device was positioned at the level of the cervicothoracic junction just below the C7 spinous process (Figure 1). Data output for the device included the percent of operative time spent “upright” and “slouched”, as well as the duration of the recorded session. Device settings were standardized for each surgeon for all cases. The threshold for “slouching” was set at 30 degrees of flexion from a neutral spinal axis. The threshold set for “slouching” in our study was 30 degrees from the neutral spinal axis. Studies conducted with the head positioned at 30 degrees anteriorly beyond the neutral position found that there is a fourfold increase in weight observed in the cervical spine, which equates to a relative risk of 2 for the development of neck pain [17,18]. The device was re-calibrated prior to each case, with the surgeon standing at a neutral, upright position as a baseline. The data from the device were sent to the surgeon’s mobile device and then uploaded to a data collection spreadsheet for analysis.

The surgeon, duration of surgery, robotic use, helmet type, consecutive case order number, and patient demographics for each case were recorded. Patient demographics collected included age, BMI (body mass index), and ASA (American Society of Anesthesiologists) classification. Two helmet types were included in the study. The Stryker Flyte surgical helmet, as well as a newer/lighter version of the Stryker T7 helmet, were studied (Stryker, Kalamazoo, MI, USA). Both helmets have a cable attaching the helmet to the battery pack below. The robot used in all robotic cases was the Stryker MAKO robot (Stryker, Kalamazoo, MI, USA). The type of implants used in the cTKAs was surgeon-dependent.

2.2. Demographics

All 103 patients were included in the final analysis. There were no exclusions. The average age of the cohort was 66.8 years and showed no difference between the robotic and conventional groups (p = 0.980). The average BMI was 31.2 and showed no difference between groups (p = 0.863). There were no differences in sex or ASA classification between groups either (

Table 1. Demographic comparisons based on robotic TKA (controlling for surgeon).

	Robotic TKA n = 48	Conventional TKA n = 55	p-Value
Age (SD)	66.8 (7.7)	66.8 (8.4)	0.980
Women % (n)	60.4 (29)	61.8 (34)	0.499
BMI (SD)	31.1 (5.0)	31.2 (4.3)	0.863
ASA Classification % (n)
1	6.3 (3)	0.0 (0)	0.560
2	41.7 (20)	37.0 (20)
3	52.1 (25)	63.0 (34)

TKA, total knee arthroplasty; BMI, body mass index; ASA, American Society of Anesthesiologists.

).

2.3. Controlling for Surgeon Posture

To control for certain differences or tendencies in posture between different surgeons, we conducted comparisons between groups clustered by surgeons. Table 1 shows baseline posture and operative time differences between the three surgeons while performing primary TKA. Posture data were also collected during clinic days to further act as a control variable for surgeon posture. Device time was manually started by the surgeon at the beginning and end of their clinic day. One surgeon had faster operative times than the other two (52.4 vs. 74.5 and 82.4 min, p < 0.001). The two other surgeons showed differences in posture throughout all TKA cases (slouch percentage, 33.2% vs. 70.9%, p < 0.001). One surgeon had shorter clinic time (158.7 vs. 287.5 and 455.7 min, p = 0.009), but all surgeons had similar postures throughout their clinic day (

Table 2. Comparison of posture between surgeons.

Variable (SD)	Surgeon 1	Surgeon 2	Surgeon 3	p Value
TKA Primary	n = 31	n = 67	n= 5
Upright %	66.8 (17.8)	29.1 (18.0)	51.8 (18.0)	<0.001
Slouch %	33.2 (17.8)	70.9 (18.0)	42.8 (18.0)	<0.001
Procedure Time (min)	74.5 (14.6)	82.4 (20.0)	52.4 (8.6)	0.001
Clinic Values	n = 4	n = 15	n = 3
Upright %	54.8 (6.2)	66.1 (25.0)	75.3 (37.5)	0.301
Slouch %	45.2 (6.2)	33.9 (25.0)	24.7 (37.5)	0.301
Device Time (min)	287.5 (166.7)	455.7 (88.2)	158.7 (30.4)	0.009

TKA, total knee arthroplasty.

). Demographic, posture, and surgical comparisons were clustered by surgeons to account for these differences.

2.4. Data Analysis

Univariate comparisons assessing differences between robotic and conventional TKA (Table 1 and

Table 3. Posture and surgical comparisons based on robotic TKA (controlling for the surgeon).

	Robotic TKA n = 48	Conventional TKA n = 55	p-Value
Posture (SD)
Upright %	57.6 (22.1)	27.5 (17.5)	<0.001
Upright (Min)	44.9 (17.4)	21.0 (14.8)	<0.001
Slouch %	42.4 (22.1)	72.5 (17.5)	<0.001
Slouch (Min)	35.4 (22.9)	54.7 (20.7)	0.037
Procedure Time (SD)	80.3 (17.2)	77.0 (21.0)	0.430
Case Number % (n)			0.711
1–3	70.8 (34)	72.7 (40)
4+	29.2 (14)	27.3 (15)
Helmet % (n)			0.198
Original	0.0 (0)	9.1 (5)
New/Lightweight	100.0 (48)	90.9 (50)

TKA, total knee arthroplasty.

) were conducted using independent t-tests for continuous variables and Chi-square tests of Independence for categorical variables. Univariate comparisons assessing differences between surgeons (Table 2) were conducted using Kruskal–Wallis tests for continuous variables and Fisher Exact tests for categorical variables. All univariate comparisons between groups were conducted using Generalized Estimating Equations (GEE) in order to cluster by surgeon. Values reported are means (Standard Deviations) for continuous variables and percentages (Counts) for categorical variables.

Multivariate linear regressions assessing predictors of slouching (% of time and total minutes) were conducted using GEE to cluster by surgeon. All analyses were conducted using two-tailed tests with an alpha set at p = 0.05 with SPSS version 27 (IBM, Inc., Chicago, IL, USA). A post-hoc power analysis was conducted using two-tailed tests of independent means, with an alpha = 0.05, and the achieved power was 0.99.

3. Results

There was a decrease in the percentage and duration of time spent slouching in raTKA cases compared to cTKA cases (42.4 vs. 72.5%, p < 0.001, 35.4 vs. 54.7 min, p = 0.037). We found no differences in operative time between the groups (80.3 vs. 77.0 min, p = 0.430). No difference was found regarding the time of day that the case took place (p = 0.711). The new/lightweight helmet was used in 100% of raTKA cases and 90.9 in cTKA (p = 0.198) (Table 3).

In the multivariate analysis, the use of robotic assistance was the only operative factor found to be an individual contributor to decreased slouching percentage and total time spent slouching (B = −28.9, −21.5, p < 0.001). In addition to the use of robotic assistance, decreased operative time was found to be an individual contributor to decreased total time spent slouching (B = 0.73, p < 0.001). Robotic assistance and lower surgical time were protective of surgeon slouching.

Patient factors that resulted in decreased slouch percentage were increased patient age and ASA 2 classification (B = −0.28, p < 0.001 and −7.17, p < 0.001). Increased patient age and ASA 2 classification also were significantly correlated with decreased total time spent slouching (B = −0.25, p = 0.048, B = −6.49, p = 0.009). These two factors were found to be protective against surgeon slouching.

The gender of the patient had no significance on slouching time or percentage of cases spent slouching (p = 0.402, 0.198). BMI was also found to have no significance on surgeon ergonomics in the multivariate analysis (p = 0.759, 0.529) (

Table 4. Multivariate linear regression predicting slouching (% and Total Time) for TKA, evaluating demographics and robotic TKA, clustered by surgeon (N = 103).

	Predicting Slouching %			Predicting Total Slouching (min)
	B	95% CI	p-Value	B	95% CI	p-Value
Age	−0.28	−0.5, −0.1	0.009	−0.25	−0.5, −0.0	0.048
Female Gender	2.15	−2.9, 7.1	0.402	2.84	−1.5, 7.2	0.198
BMI	0.08	−0.4, 0.6	0.759	0.14	−0.29, 0.56	0.529
ASA 2 Classification	−7.17	−12.4, −2.0	0.007	−6.49	−11.4, −1.6	0.009
Procedure Time	0.10	−0.1, 0.3	0.388	0.73	0.64, 0.83	<0.001
Robotic TKA	−28.9	−40.9, −17.0	<0.001	−21.5	−29.9, −13.2	<0.001

TKA, total knee arthroplasty; BMI, body mass index; ASA, American Society of Anesthesiologists.

).

4. Discussion

Orthopedic surgery is a labor-intensive field that places great stress on the musculoskeletal system through repetitive tasks performed in non-ergonomic positions. The neck and lower back are especially at risk due to poor posture in the operating room [3,4,5,6,7,8]. Among the subspecialties of orthopedics, adult reconstruction may be the most laborious [6,7,9,10]. Robotic-assisted technology is becoming an increasingly utilized adjunct within the realm of primary TKA [12,13]. Not only does robotic surgery offer reliable bony preparation and similar patient satisfaction to cTKA, but a recent study showed decreased surgeon cardiovascular stress and strain in raTKA versus cTKA [16]. Our multivariate analysis found an ergonomic benefit for surgeons when using the robot for primary TKAs versus performing the procedure manually.

In a recent cadaveric study using motion sensors on the T3 and occiput to compare cervical spine postures and repetitive motions in surgeons performing conventional and robotic TKA, Scholl et al. found significant differences in cervical and thoracic ergonomics [19]. They found that performing raTKA reduced repetitive motion and time spent in non-neutral positions in the cervical spine compared to cTKA. Our in vivo study supports their results, as we found that the use of the robot was a significant individual predictor of decreased slouching percentage. On average, the use of the robot decreased surgeon slouching time by 19.3 min (26.6%, p < 0.001). Operative time was an individual contributor to time spent slouching, but not a percentage. In another study, Meltzer et al. observed deviation from the neutral spinal axis in a single orthopedic surgeon using motion sensors in a series of cases. They found that the surgeon spent over 50% of the cases in high-risk neck and torso positions [20]. We found similar results in our study as the surgeons spent 72.5% of the case slouched in cTKA versus only 42.4% in raTKA (p < 0.001).

In our study, we attempted to quantify surgeon posture and identify factors that may help or hinder “slouching” posture in the operating room while robotic versus conventional TKA. Similar studies have been conducted in hand and spine surgery using the Upright Go system to provide postural feedback. Johnson et al. conducted a study observing three hand surgeons wearing the UpRight Go device. They found a 40% slouch rate in surgeons performing all types of hand and upper extremity cases [21]. This was similar to the slouch rate found in robotic TKA. Kothari et al. found similar slouching rates in spine surgeons unless the surgery involved deformity correction. Adult deformity cases slouch rates of 60%, which was closer to the rates we found in conventional TKA [22]. These studies help to validate the accuracy of this device for measuring surgeon posture intraoperatively. To show the validity of this device, Harvey et al. performed a study showing the positive effects of posture feedback training on personal health using the Upright Go device. The postural feedback group showed improvements in energy, mood, and self-reported neck and lower back pain compared to the control group. This study provides insight into the clinical utility of this device [23].

The development of robotic technologies in other surgical fields has promise for improving surgeon posture. A gynecologic surgery study of robotic use in hysterectomies showed both better ergonomic and lower cardiovascular demands in robotic surgery versus conventional [24]. Similarly, a urologic study in robotic-assisted radical prostatectomies showed a reduction in demanding neck postures by 24% for the performing surgeon while using the robot [25]. Our study supports these findings on ergonomic improvement with the use of robotic technology.

In a recent study of 40 primary TKA by a single surgeon, Haffar et al. found that raTKA results in less surgeon physiological stress and energy expenditure compared to cTKAs [16]. Their study also found significant increases in the surgeon’s cervical and lumbar flexion in conventional versus raTKA, indicating poorer posture in the conventional group. These findings agree with our study; however, their study only included data from a single surgeon who was self-described as a novice in the robotic system they used during that study. In contrast, the current study included surgeon posture data during 103 TKAs performed by three surgeons while controlling for individual surgeon posture. Furthermore, each surgeon was experienced in the robotic and conventional methods included during this study, making the results of our study more accurate in regard to the ergonomic strain seen during a surgeon’s normal surgical workflow.

There were several limitations to our study. First, while we prospectively collected data, patients were not randomized to receive conventional or robotic techniques. Therefore, patient preferences for robotic surgery or surgeon selection bias for the use of robotic assistance may have skewed our results. Second, only one robotic system was used during this study, which does not allow us to generalize our findings across other robotic platforms. Third, surgeons in this study knew that they were being studied, which led to potential bias through the Hawthorne effect. No subjective data from the surgeon was collected at the end of the case. The surgeons included did not report any injuries or missed time during this study period. This data could have helped support the idea that better posture (keeping within 30 degrees of upright) aids in decreasing surgeon pain. If posture analyses in ergonomics were evaluated with REBA (Rapid Entire Body Assessment) or RULA (Rapid Upper Limb Assessment), clearer results could be achieved. Lastly, the case numbers between the three surgeons varied widely depending on the availability of the surgeon during the study period, with only one posture device shared amongst surgeons. Therefore, a single surgeon’s postural tendencies may have skewed our data despite controlling for the surgeon. Despite these limitations, this is the largest and most diverse series of in vivo surgeon posture comparisons between conventional and robotic total knee arthroplasties.

5. Conclusions

Surgeons performing primary robotic-assisted TKA cases spend significantly less case time and case percentage in a slouched posture compared to primary cTKA cases. To our knowledge, this is the largest study observing the ergonomic benefit of robotic-assisted technology in primary TKA. Further research is needed to determine the long-term effects of orthopedic surgeon posture on pain and career longevity.