Return to Work After Autologous Chondrocyte Implantation of the Knee in Patients with Workers Compensation Insurance

Abstract

Objective

This retrospective analysis aimed to evaluate patient and defect characteristics on return to work after autologous chondrocyte implantation (ACI) for all patients with workers compensation insurance in our clinic.

Methods

Retrospective analysis of medical records of patients with workers compensation treated with ACI in our clinic over 10 years (August 2004 to November 2014). Data were collected on demographics, lesion size and location, number of defects, duration of symptoms, and outcomes.

Results

A total of 28 patients with 30 ACI procedures were included in this analysis. Patients had a mean age of 40.0 years, were mostly male, and tended to be overweight (mean body mass index = 31.5 kg/m²). Most patients were employed in occupations with high physical demand (85.7%) compared with lower physical demand jobs (14.3%). The mean number of lesions per knee = 1.73. Overall, the mean total surface area of the defect was large (12.6 cm²), with large mean individual lesion size of 7.39 cm². All 28 patients returned to work in some capacity, with a total of 22 (79%) returning to full work, and 6 (21%) returning to modified work. There were no significant differences in the comparison of patients who returned to full versus modified work, in mean age, body mass index, high versus low physical demand occupation, or lesion characteristics (size, chronicity, or location).

Conclusions

Our results demonstrate that in a challenging workers compensation cohort, in patients with multiple large defects, treatment with ACI allows a high percentage of patients to return to their normal vigorous work activities within 9 months to 1 year following cartilage restoration. This study is a level IV retrospective case series.

Introduction

Autologous chondrocyte implantation (ACI) for cartilage repair of the knee has been associated with satisfying patient outcomes.^1-3 However, limited studies describe return to work (RTW) after ACI,^4,5 and few studies evaluate patients with workers compensation–related knee injuries treated with ACI. ⁶ Historically, patients with workers compensation injuries are associated with poor outcomes after surgical procedures, and for this reason, many surgeons are reluctant to operate on this group. A systematic review of 20 studies by de Moraes et al. ⁷ showed that the overall risk for unsatisfactory results after orthopedic surgery for patients with workers compensation injuries compared with nonwork-related injuries is 2.08 (95% confidence interval 1.54-2.82).

In a study by Yates, ⁶ 24 patients treated with ACI and with workers compensation insurance were studied. The average size lesion was 4.74 cm², and 75% (18/24) had physically demanding jobs. Of the 24 patients, 13 (54%) returned to full work with no restrictions at 1-year follow-up. In patients with >1-year follow-up, 62% (13/21) returned to unrestricted work status.

The purpose of our research is to report the outcomes and RTW rate in patients treated with ACI at our clinic for over 10 years. Patients had workers compensation–related injuries of the knee with most having vigorous occupational requirements, including firefighters, police officers, border patrol and immigration officers, and jobs in trades such as electrician, tile setter, and painter. We sought to evaluate the effects of patient demographics, cartilage defect characteristics, and concomitant pathology with concomitant procedures on RTW after ACI.

Methods

Patient Cohort

Institutional review board approval was obtained before the collection of data for this retrospective study. We performed a retrospective analysis of medical records of 29 patients (31 knees) treated with ACI at our clinic over 10 years between August 2004 and November 2014 with work-related injuries. During data review it was found that 1 patient had a wrist injury that precluded him from returning to work; therefore, this patient was excluded from the RTW analysis for this article. The medical records examined included operative reports and follow-up visit forms. The operative notes contained detailed information about the cartilage defect and operative procedures. Follow-up visit forms included range of motion, time from surgery, clinical progress, and work status. Duration of time from ACI surgery to return to work was calculated from the date of ACI surgery to the date the patient returned to both modified and full work. Exclusion criteria included lack of workers compensation injury. We report patient demographics and their correlations with regard to surgical outcomes following ACI surgery as determined from patient follow-up visit reports and changes in patient-reported outcomes in the Knee Injury and Osteoarthritis Outcome Score (KOOS) and International Knee Documentation Committee (IKDC) scores. We report the number of patients who returned to work, mean duration of time to return to modified and full work, and reasons for not returning to work. The effect of concomitant procedures on RTW is also discussed in this analysis. Return to full work was defined as a return to unrestricted activity at the same job and employer. Return to modified work was defined as a return to reduced activity level at the same job, or a return to a job with a different employer.

Surgical Technique

Two-stage articular cartilage restoration was performed with the initial stage involving articular cartilage harvesting and subsequent cell expansion. Some patients underwent concomitant procedures such as revision anterior cruciate ligament (ACL) reconstruction, osteotomy, or other procedures. During the secondary procedure when the ACI was implanted, the technique was as noted by Peterson et al. ¹ and Minas et al. ⁸ If necessary, concomitant procedures such as proximal tibial osteotomy, tibial tubercle osteotomy, and meniscal allografts were performed. During the second stage of surgery, an open arthrotomy was performed exposing the chondral defect on the affected articular surface, and a scalpel and ring curette were used to debride the cartilage and create stable vertical shoulders. The calcified cartilage was debrided off of the subchondral bone while preserving the integrity of the subchondral bone. A template was created, and this was used to fashion the graft overlying the defect. For first-generation ACI (ACI-P) procedures, periosteum was harvested from the proximal tibia or distal femur, placed on the defect and sutured using a 6-0 Vicryl suture. For second-generation ACI (ACI-C), a collagen I/III membrane (Geistlich, Germany) was used to cover the defect instead of periosteum. Twelve million chondrocytes were extracted from each transport vial and were resuspended with an 18-gauge needle. The chondrocytes were then injected under the periosteal/collagen patch that had been sutured to the surrounding intact articular surface to cover the surface defect through a small opening, and the injection site was subsequently sutured closed in a similar fashion. Then, the suture line of the periosteum or membrane was sealed with fibrin glue. Concomitant procedures such as osteotomies and meniscal allografts were performed prior to implantation of the membrane and cells to decrease the risk of disruption of the graft with leakage of the underlying chondrocytes that were suspended in a liquid suspension.

Clinical Outcome Measurements

Outcome measurements included the patient-reported outcomes (PROs) of KOOS and IKDC. Forms were collected preoperatively and at follow-up for patients. Follow-up in this patient population was difficult and only a subset of patients completed documentation.

Postoperative Care

Postoperatively, the knees were immobilized in full extension for 8 hours and then started on a continuous passive motion machine (CPM). The range of motion depended on the area of the knee which underwent cartilage resurfacing. Rehabilitation after surgery involved a continuous passive motion for 6 to 8 hours a day for 6 weeks. Partial weightbearing occurred for 6 weeks, and progression to full weightbearing occurred between 6 to 12 weeks dependent on whether the patient underwent an osteotomy procedure. For patellar and trochlear lesions, the range of motion was limited to 0° to 30° in the first month in the CPM. Each day, the patient would use gravity assistance to flex to 90° passively to stretch the quadriceps. For lesions in other locations, the goal was to achieve 90° of flexion within one month postoperatively. The ultimate goal of the rehabilitation was to protect the ACI graft and promote gradual and safe return to function and activity through the 4 phases of rehabilitation: (1) the early protection phase (phase I—weeks 0-6), (2) the transition phase (phase II—weeks 6-12), (3) remodeling phase (phase III—weeks 12-26), and (4) maturation phase (phase IV—weeks 26-52). In order to match the growth and maturation of the cellular graft, we delayed a return to full activities for 12 months in many cases given the vigorous job activities of lifting, twisting, running, and pivoting, which the patients performed.

Statistical Analysis

Microsoft Excel was used for data entry, management, and statistical analysis was conducted by using SAS version 9.2 (SAS Institute, Cary, NC). Patient data were de-identified. Summary statistics included the mean, standard deviation, and median for the patient and lesion characteristics and return to work duration. The T test was performed to compare means of patient and knee characteristic variables for the return to full versus modified work duration between the following patient groups. Also, a regression analysis was performed to evaluate relative risk and confidence intervals for a limited number of variables. The null hypothesis was that there was no statistically significant difference in return to work duration between these patient groups. The level of statistical significance was determined by a P ≤ 0.05. Graphs were generated in Microsoft Excel or GraphPad Prism (San Diego, CA) to illustrate trends in mean patient characteristics, PROs and mean duration for a return to work.

Results

Patient Characteristics

A total of 28 patients with 30 ACI procedures were included in this analysis. As shown in Table 1 , patients in the study had a mean age of 40.0 years (range 23-51 years), were mostly male (22 male, 6 female), and tended to be overweight (mean body mass index [BMI] = 31.5 kg/m² [range 20.3-48.7 kg/m²]). Among all patients, most were employed in occupations with high physical demand (85.7%) that included civil/response, trades, and sports, compared with lower physical demand jobs (14.3%) that included sales and service.

Table 1.

Patient Characteristics.

	Total	Return to Work
		Return to Full Work	Return to Modified Work
By patient summary	N = 28 a	N = 22 b (79%)	N = 6 (21%)
Gender
Male	22 (78.6%)	18	4
Female	6 (21.4%)	4	2
Age (years)
<30	2 (7.2%)	2	0
30-40	11(39.3%)	8	3
41-50	14 (50.0%)	11	3
>50	1 (3.6%)	1	0
Mean (SD)	40.0 (5.9)	40.0 (6.6)	40.6 (3.1)
Median (range)	41 (23-51)	42.5 (23-51)	41 (34-46)
Body mass index (kg/m2)
18.5-24.9	4 (14.3%)	4	0
25.0-29.9	9 (32.1%)	5	4
30.0-39.9	11 (39.2%)	10	1
≥40	4 (14.3%)	3	1
Mean (SD)	31.5 (5.3)	31.9 (5.6)	30.5 (3.8)
Median (range)	30.2 (20.3-48.8)	30.7 (20.3-48.8)	29.3 (25.1-40.6)
Occupation
High physical demand	24 (85.7%)	19 (79.2%)	5 (20.8%)
Civil/Response c	13	10	3
Trades d	9	7	2
Sports e	2	2	0
Lower physical demand	4 (14.3%)	3 (75.0%)	1 (25%)
Sales and service f	4	3	1

^aExcludes 1 patient from the analysis because they had a wrist injury that precluded them from returning to work.

^bOne patient was counted only once (as returned to full work) following their first autologous chondrocyte implantation surgery; however, the patient had a second surgery two years later and returned to modified work (not counted in by-patient summary). The 2 knees are counted separately in the by-knee characteristics in Table 2 .

^cCivil/Response: Transportation Security Administration (TSA) agent, firefighter (2), firefighter/paramedic, a police officer (3), border patrol, immigration officer, security guard, sheriff, corrections officer, a psychiatric nurse.

^dTrades: HVAC (heating, ventilation, and air conditioning) technician, tile setter, electrician, painter, bus driver, truck driver, lead mechanic; superintendent; restaurant manager.

^eSports jobs: soccer coach, golf pro.

^fSales and service: clerk, event planner, interior designer, cashier.

All 28 patients in the study returned to work in some capacity, with a total of 22 (79%) returning to full unrestricted work, and 6 (21%) returning to modified work or a different employer/job.

One of the 22 patients returning to unrestricted work one was converted to total knee arthroplasty 7 years after he had returned to work.

There were no significant differences between mean age (P = 0.84), mean BMI (P = 0.26), or gender (0.42) between patients who returned to full work and those who did not return to their previous job activities (modified work). There were no significant differences between high or low physical demand occupations in return to full or modified work (P = 1.0).

Knee Characteristics

As shown in Table 2 , 30 knees were treated with ACI and evaluated for RTW. Overall, the time from injury to treatment (chronicity) was 3.59 years (range 6 months to 13.3 years). The majority of knees (63%) had at least 2 lesions with the mean number of lesions per knee = 1.73. The most common lesion locations, found in almost half of all knees, were on the trochlea/medial femoral condyle (TRO/MFC) or MFC alone (14/30 knees total; 46.7%). Overall, the mean total surface area of the defect was very large (12.6 cm²), with a large mean individual lesion size of 7.39 cm². There were no significant differences between mean chronicity (P = 0.228), mean defect size total surface area (P = 0.39), or mean lesion size (P = 0.63) for the 23 knees who returned to full work compared with the 7 knees who did not return to their previous job activities (modified work).

Table 2.

Knee Characteristics.

	Total	Return to Work
		Return to Full Work	Return to Modified Work
By-knee summary	N = 30 a	N = 23	N = 7
Chronicity (years)
<1	4 (13.3%)	3	1
1-2.9	13(43.3%)	11	2
3-6	8 (26.7%)	5	3
6-14	5 (16.7%)	4	1
Mean (SD)	3.65 (2.45)	3.4 (2.4)	4.7 (2.6)
Median (range)	2.5 (0.52-13.3)	2.4 (0.5-13.3)	4.6 (0.9-12.5)
Lesion location and total defect size surface area; mean number of lesions per knee = 1.73
TRO/MFC	7 (33.3%)	6	1
MFC	7 (16.7%)	5	2
MFC/PAT	3 (10.0%)	3	0
TRO	3 (10.0%)	2	1
PAT	2 (6.7%)	1	1
TRO/MFC/PAT	2 (10.0%)	2	0
TRO/MFC/LFC	1 (3.3%)	1	0
TRO/PAT	1 (3.3%)	1	0
TRO/PAT/LFC	1 (3.3%)	0	1
TRO/MFC/Tibial plateau	1 (3.3%)	1	0
LFC	1 (3.3%)	1	0
TRO/LFC	1 (3.3%):	1	0
Mean (SD) total surface area, cm2	12.6 (6.2)	13.1 (6.3)	10.8 (5.4)
Median (range) total surface area, cm2	10 (4-33)	11.5 (4-33)	10 (4-22)
Individual lesions by location and mean lesion size
MFC	20	16	4
TRO	17	14	3
Patella	9	6	3
LFC	4	3	1
Tibial plateau	1	1	0
Mean (SD) lesion size, cm2	7.39 (3.14)	7.53 (3.2)	6.9 (2.98)
Median (range) lesion size, cm2	6.25 (1.5-20)	6.5 (1.5-20)	5.63 (2.25-12)

LFC = lateral femoral condyle; MFC = medial femoral condyle; PAT = patella; TRO = trochlea.

^aExcludes one knee from the analysis because the patient had a wrist injury that precluded them from returning to work.

Duration of Return to work

Among all 28 patients, there were data for 20 patients for the time of return to partial work (i.e., patients were under some restrictions for returning to work), regardless of whether they eventually returned to full or modified work. For these 20 patients, the mean duration from ACI surgery to return to partial work was 4.1 months. For the 22 patients who returned eventually to full work, the mean duration from ACI surgery to return to full work was 12.4 months (4 patients ≤9 months, 12 patients >9 to ≤12 months, and 6 patients >12 months).

A Concomitant Procedure Does Not Significantly Affect the Time to Return to Work

Among all knees, 50% (15/30) of knees underwent concomitant procedures with ACI surgery including tibial tubercle osteotomy (TTO), high tibial osteotomy (HTO), ACL reconstruction, or meniscal allograft. Of the 15 knees with concomitant surgeries, 12 (80%) returned to full work and 3 (20%) returned to modified work. Of the 15 knees without concomitant surgery, 11 (73%) returned to full work and 4 (27%) returned to modified work.

Among the 22 patients (23 knees) that returned to full work the mean duration for RTW was similar in patients with knees that required concomitant procedures (n =12 knees) and those that did not require concomitant procedures (n = 11 knees; P = 0.89) ( Fig. 1 ).

Figure 1.

Effect of Concomitant procedures on return to work in patients who returned to full work. Mean duration of return to full work was similar in patients with knees requiring a concomitant procedure and those that did not.

Clinical Outcomes

KOOS scores were collected preoperatively from 15 patients, at 1-year postop (n = 5), and 2 years (n = 4). The IKDC was collected preoperatively from 9 patients and at 2 years (n = 5), or last visit follow-up >2 years from 7 patients. As we collected these outcomes on only a few of our patients, we are unable to make rigorous statistical statements. Our data demonstrated that there was a general increase in scores over time for each of the KOOS subscales ( Fig. 2 ) and IKDC scores ( Fig. 3 ) signifying an improvement in function and symptoms.

Figure 2.

KOOS subscores (mean and SEM) for patients with baseline (n = 15, white fill), 1 year (n = 5, gray fill), and 2 years (n = 4, black fill) postsurgery follow-up data. KOOS, Knee injury and Osteoarthritis and Outcome Score; SEM, standard error of the mean; ADL, function in activities of daily living; QOL, knee-related quality of life; Sympt, other symptoms; Sport, function in sport and recreation.

*Indicates a statistically significant difference (P < 0.05; repeated-measures analysis of variance comparison of visit [baseline, year 1, and year 2]).

Figure 3.

International Knee Documentation Committee (IKDC) scores (mean and SEM) for patients with baseline, 2-year, and last visit postsurgery follow-up data.

*Indicates a statistically significant difference (P < 0.05; repeated-measures analysis of variance comparison of visit [baseline, year 2, and last visit]).

Discussion

We can return an active population of workers compensation patients with complicated knee injuries who perform vigorous work activities back to work approximately 1 year after ACI resurfacing. Our study demonstrated that 79% of patients who sustained work-related injuries treated with ACI and additional concomitant procedures returned to work within approximately 1 year of their graft implantation. Most of these patients came from occupations with high physical demands, including firefighters and police officers. Historically, patients with workers compensation injuries are associated with poor outcomes after surgical procedures, and for this reason, many surgeons are reluctant to operate on this group. The specific reasons for poor results remain difficult to determine, but psychosocial factors are implicated, including stress from economic factors. A study looking at patient-reported factors influencing return to work after joint replacement ⁹ found that RTW is influenced by patient clinical and occupational factors. Some of these factors included workplace support, job role adaptation, and improved physical and psychological performance after surgery compared to preoperative status.^1,8,9 A systematic review of 20 studies by de Moraes et al. ⁷ showed that the overall risk for unsatisfactory results after orthopedic surgery for patients with workers compensation injuries compared with nonwork-related injuries is 2.08 (95% confidence interval 1.54-2.82).

There has been 1 previous study that evaluated RTW in patients with workers compensation injuries that were treated with ACI for isolated cartilage defects. Yates ⁶ evaluated the effectiveness of ACI in the treatment of 24 patients with workers compensation insurance. Of the 24 patients, 13 (54%) returned to full work with no restrictions at 1-year follow-up. In patients with >1-year follow-up, 62% (13/21) returned to unrestricted work status. While the mean age of 38 years in the Yates study was similar to our population, differences between the Yates study and our study include the following: In the Yates study a smaller lesion size (4.7 cm² vs. 7.4 cm²), a single lesion was present per knee, shorter duration of symptoms prior to ACI, and no additional significant pathology. In our patients, 50% of knees underwent additional surgery with 40% undergoing osteotomies or meniscal allograft treatment as well. In spite of the complexity of the injuries in patients in our study, we were able to return a greater percentage of patients to full work in a vigorous work capacity compared with those noted by Yates.

Additionally, Minas et al. ⁸ reported minimum 10-year outcomes of 200 patients who had been treated with ACI, demonstrating overall survivorship of 71% and improved function in 75% of patients at 10 years. A subanalysis of 46 patients who had workers compensation insurance showed no significant difference in survivorship between these patients and the rest of the cohort. Our data would be more similar to the results noted by Minas et al. in the subgroup in their study. However, length of time to return to work ⁸ after ACI surgery was not discussed.

There have been 2 additional studies that evaluated RTW in patients who have undergone ACI resurfacing in a non–workers compensation population. Pestka et al. ⁵ reported an RTW in 130 patients treated with ACI with a mean patient age of 36.2 years and a mean defect size of 4.4 cm². The study found that work intensity influenced the return to work rates and duration of absenteeism after surgery with workplace adaptations being required in 9.2% of the cases.

Bode et al. ⁴ reported a study of 40 patients who underwent concomitant HTO and ACI for full-thickness cartilage defects of the MFC and also found that mean time off of work depended on workload. The effect of concomitant procedures, particularly osteotomies on return to work has been previously reported in a general population.^10,11 A systematic review of 26 studies evaluated return to sports and RTW on 1321 patients at 4.8 years of follow-up ¹⁰ in patients who had undergone knee osteotomies. Of the 14 studies that reported RTW, the rate was 85%. A systematic review by Ekhtiari et al. ¹¹ evaluated the timelines for a return to sport and work of over 1189 patients from 19 studies who had undergone HTOs. Overall, 84.5% of patients returned to work postoperatively, but only 65.5% returned at an equally functional level which is less than our improvement and may suggest that the addition of the ACI improves the functional capacity after HTO.

Pike et al. ¹² reported a retrospective case series of 26 patients with large defects with a mean size 8.4 cm² who were treated concomitantly with ACI and an ACL reconstruction. Patient outcomes were a minimum of 2 years and an average of 8 years. Half of the patients (n = 13) had single lesions, and 13 patients had multiple defects of which 5 patients had bipolar lesions. There were statistically significant improvements in clinical outcomes, including modified Cincinnati Knee Rating System, Western Ontario and McMaster Universities Osteoarthritis Index, and visual analogue scale from baseline to follow-up (P < 0.0001). Also, a majority of the patients (69%) reported an improvement and rated their knee as good or excellent. Comparison of outcomes between patients with single and multiple lesions showed no statistically significant difference.

Usually, treatment of patellofemoral defects is challenging due to the biomechanical complexity of the joint, and microfracture outcomes of the patella have been poor. ¹³ There have been studies reporting outcomes for ACI in the patella with 20 years of follow-up with more than 90% patient satisfaction. Peterson et al. ¹ had shown a greater than 70% success rate with 20 years of follow-up when malalignment was concomitantly addressed. Others have also shown good outcomes in the patella with high patient satisfaction rates.^14-16 In our case series, 30% (9/30) knees had a patella lesion. Six (67%) of the 9 patients with patella defects returned to full work, and 3 (33%) returned to modified work. When combined with TTO, ACI in the patella has produced better outcomes of 80% to 85% whereas when without TTO outcomes have ranged from 40% to 55%. ¹⁷

The treatment of trochlear lesions presents a challenge due to the topography of the joint and healing may be impaired due to high shear forces. The case series by Mandelbaum et al. ¹⁸ reported outcomes after treatment of trochlea lesions in 40 cartilage repair registry patients, of whom 43% were receiving workers compensation benefits. The patients were followed for a mean of 59 ± 18 months using the modified Cincinnati Knee Rating System. The mean lesion size was 4.5 cm². There was a statistically significant improvement in pain (2.6-6.2 points), swelling (3.9-6.3 points), and overall condition (3.1-6.4) according to the modified Cincinnati Knee Rating System, and there were no failed implantations. In our study, 57% (17/30) of the knees presented with trochlear lesions. Of the patients with trochlear lesions, 82% (14/17) returned to full work. One patient had trochlear lesions on both the left and right knee that were treated with ACI on 2 separate occasions. This patient returned to full work. There were 18% (3/17) that returned to modified work. One of these patients underwent bilateral staged procedures with TTOs.

BMI is considered an important factor in treatment decisions for cartilage defect ¹⁹ and on ACI outcomes. ²⁰ A BMI of 30 kg/m² is considered an upper limit for cartilage repair and it is believed that high BMI is associated with poorer outcomes after cartilage repair. ²¹ In our case series, the mean BMI was 31.6 kg/m², which is higher than the upper limit considered optimal for cartilage repair. The mean BMI of the 22 patients who returned to full work was 31.9 kg/m², and the mean for those that returned to modified work was 30.5. There was no statistical significance between the BMI between the 2 groups, and our studies demonstrated good RTW in spite of high BMI.

ACI is indicated for patients aged 18 to 55 years old. It is believed that cell therapy performs better in younger patients. However, reports that have evaluated ACI outcomes in older patients have demonstrated good to excellent results. The mean patient age in our study was 40 years and ranged from 23 to 51 years. The mean age of the patients who returned to both full and modified work was the same.

The limitations of our study are the small sample size, the lack of a control group, and a limited collection of patient-reported outcomes. We previously collected some PROs for patients on paper, and enrollment was inconsistent. With the availability of electronic registries such as Surgical Outcomes System and the International Cartilage Repair Registry, we hope to follow our current and new patients prospectively.

Conclusion

Our results demonstrate that in a challenging workers compensation patient cohort, with multiple large defects of the knee cartilage, treatment with ACI allows a high percentage of patients to return to their normal vigorous work activities within 9 months to 1 year following cartilage restoration. Our results also demonstrate that in patients with coronal malalignment and patellofemoral lesions, treatment with ACI with concomitant osteotomies and other procedures allows a high percentage of patients to return to their normal vigorous work activities. The presence of additional risk factors in this patient population, such as high BMI did not appear to affect the return to full work rate. Consideration of joint preservation in these patients can provide a successful, durable result in spite of potential psychosocial factors related to a workers compensation status and is worth consideration.