Hostname: page-component-74656f9884-c52kr Total loading time: 0 Render date: 2026-11-07T07:36:49.414Z Has data issue: false hasContentIssue false
  • English
  • Français

Cost and health status analysis after autologous chondrocyte implantation and mosaicplasty: A retrospective comparison

Published online by Cambridge University Press:  04 August 2005

Sarah Derrett ,
Elizabeth A. Stokes ,
Marilyn James ,
William Bartlett  and
George Bentley
Sarah Derrett
Affiliation:
Keele University
Elizabeth A. Stokes
Affiliation:
Keele University
Marilyn James
Affiliation:
Keele University
William Bartlett
Affiliation:
Royal National Orthopaedic Hospital
George Bentley
Affiliation:
Royal National Orthopaedic Hospital
Rights & Permissions [Opens in a new window]

Abstract

Objectives: Chondral defects of the kneecartilage are prevalent. Autologous chondrocyte implantation (ACI) and mosaicplasty are increasingly used to treat symptomatic knee defects. This study assessed the costs and health status outcomes after ACI and mosaicplasty.

Methods: Patients were eligible to participate in this cross-sectional study if they received ACI or mosaicplasty at the Royal National Orthopaedic Hospital between 1997 and 2001 or were on a waiting list for ACI. Secondary-care resource use was collected to 2 years postoperatively using a resource collection proforma. Participants responded to postal questions about sociodemographic characteristics and knee-related (Modified Cincinnati Knee Rating System) and general health status (EQ-5D).

Results: Fifty-three ACI, twenty mosaicplasty, and twenty-two patients waiting for ACI participated. The average cost per patient was higher for ACI (£10,600: 95 percent confidence interval [CI], £10,036–£11,214) than mosaicplasty (£7,948: 95 percent CI, £6,957–£9,243). Postoperatively, ACI and mosaicplasty patients (combined) experienced better health status than those waiting for ACI. ACI patients tended to have better health status outcomes than mosaicplasty patients (not statistically significant). Estimated average EQ-5D social tariff improvements for quality-adjusted life year (QALY) calculations were 0.23 (ACI) and 0.06 (mosaicplasty). Average costs per QALY were £23,043 (ACI) and £66,233 (mosaicplasty). The incremental cost effectiveness ratio (ICER) for providing ACI over mosaicplasty was £16,349.

Conclusions: Average costs were higher for ACI than mosaicplasty. However, both the estimated cost per QALY and ICER for providing ACI over mosaicplasty fell beneath an implicit English funding threshold of £30,000 per QALY. Prospective studies should include measures of utility to confirm the estimated cost utility ratios of ACI and mosaicplasty.

Keywords

Autologous chondrocyteMosaicplastyCost effectivenessOutcomeKnee

Information

Type
GENERAL ESSAYS
Information
International Journal of Technology Assessment in Health Care , Volume 21 , Issue 3 , July 2005 , pp. 359 - 367
Copyright
© 2005 Cambridge University Press

Chondral defects of the knee cartilage are prevalent in young populations. A retrospective review of over 31,000 arthroscopies in the United States found chondral lesions in 63 percent of patients (13). Damaged knee articular cartilage may result in pain, functional impairment, and symptoms including joint locking, swelling, and giving way (3;20), and may contribute to the early onset of osteoarthritis, although natural history studies are required (10;26).

Internationally, several surgical treatments have been developed to treat chondral defects. Arthroscopic lavage and debridement of loose cartilage have been reported to provide short-term symptomatic relief when applied to small defects (11;24). Techniques including abrasion arthroplasty, subchondral drilling, and microfracture have since been developed to penetrate subchondral bone and induce repair by liberating mesenchymal stem cells (17;37;39;41;42). However, evidence of long-term efficacy is limited for many of these procedures, which commonly promote repair by softer less-durable fibrocartilage rather than the desirable hyaline cartilage (6;18;20;26;27).

Recently, new technologies have been developed with the aim of stimulating hyaline cartilage growth over the damaged articular surface (3;18). Mosaicplasty and autologous chondrocyte implantation (ACI) are two techniques increasingly used for the treatment of symptomatic osteochondral defects in the knee. Mosaicplasty began to be used clinically in 1992 (20;21). The procedure consists of transplanting a series of small cylinders of osteochondral tissue from patient self-donor sites into the lesion by open surgery or, occasionally, by arthroscopy. Mosaicplasty is recommended for lesions between 1 and 4 cm2 and for younger patients (20). Associated adverse events include deep infections, hemarthroses, and thromboembolic complications—but these complications are infrequent. Other problems include graft size limitations and donor site morbidity. Despite such difficulties, studies indicate good or excellent clinical outcomes after mosaicplasty in 69 to 94 percent of recipients (3;20).

ACI was introduced in Sweden in 1987 (7). ACI requires an arthroscopic biopsy to collect (harvest) healthy cartilage cells. The cells are then reproduced in culture—usually over 2 to 5 weeks. During a second surgical episode, cultured cells are transplanted into the debrided articular lesion. Variations of the implantation technique include the use of periosteal (8) and porcine-derived type I/III collagen membranes (3;18) to cover the chondral defect and retain cultured cells. Associated complications include graft hypertrophy, delamination, and arthrofibrosis. Re-operation rates due to complications after periosteum-covered ACI are between 16 percent and 25 percent (22;29;31). A smaller proportion of patients with collagen membrane cover experience complications (18). Studies have found good to excellent outcomes after ACI in 60 to 90 percent of recipients (3;23;26;35).

Many treatments for articular cartilage injuries are expensive. Currently, there is little evidence for the costs and effectiveness of these new treatments (8). This paper presents results from a study undertaken by a multidisciplinary team investigating the costs and health status effectiveness of ACI and mosaicplasty.

PARTICIPANTS AND METHODS

Design and Participants

Patients were eligible for this cross-sectional retrospective cohort study if they received first ACI or first mosaicplasty for chondral or osteochondral lesions of 1 cm diameter or more at the Royal National Orthopaedic Hospital (RNOH) between March 1997 and February 2001. Some participants from the two surgical groups previously had been involved in a randomized controlled clinical trial of ACI and mosaicplasty (3). A third group was approached to participate if they were on an ACI waiting list at RNOH between December 2002 and April 2003. Waiting list participants completed a postal health status questionnaire to provide information about the knee-related and general health status of preoperative patients. People were ineligible for the study if they were younger than 16 years at first referral, unable to complete the English-language postal questionnaire, had operations on both knees, or both a mosaicplasty and an ACI operation on different sites of the same knee. Participants with subsequent operations on the original knee site remained in their original ACI or mosaicplasty group for analysis. Ethical approval for this study was obtained from the RNOH NHS Trust Research and Ethical Committee.

Surgical Treatment and Rehabilitation

Patients were initially reviewed in the outpatient department. After radiographic assessment, patients clinically suitable for ACI or mosiacplasty were referred for a magnetic resonance imaging (MRI) scan of the knee and knee arthroscopy. After arthroscopic confirmation of suitability, cartilage harvesting was performed for the ACI recipients at day surgery. Occasionally, an overnight stay was required if patients had traveled some distance to the hospital. The cartilage fragment was placed into a transport medium and sent to a laboratory where the chondrocytes were cultured. After 3 to 5 weeks, the ACI patients were readmitted and ACI was performed. After surgery, the patient's knee was immobilized in a plaster-of-Paris backslab and full weight-bearing with crutches was encouraged. After 2 to 3 days, when the postoperative swelling had reduced, a light cylinder cast was applied and the patient discharged. The plaster was removed after 10 days, and a daily physiotherapy-directed rehabilitation program commenced. A full range of physical activities was not recommended during the first postoperative year. Clinical progress was reviewed in the outpatient department at 6 weeks, 12 weeks, 6 months, 9 months, 12 months, and then twice yearly. Patients who were unable to flex the knee beyond 90 degrees after 6 weeks were referred for manipulation under anesthesia (MUA) and intensive inpatient physiotherapy. After 1 year, all patients underwent arthroscopic assessment of the graft and biopsy when possible. Additional treatment arthroscopies were performed in patients who developed symptoms suggestive of graft hypertrophy or delamination.

In contrast to ACI, mosaicplasty was usually performed under the same anesthesia as the assessment arthroscopy. Mosaicplasty patients required postoperative mobilization, physiotherapy, and clinical review similar to ACI patients.

Resource Use and Costs

Secondary-care resource use related to the ACI or mosaicplasty was collected from patients' electronic and medical records at RNOH. Treatments received at other hospitals were outside the scope of this study. Information was collected about presurgery arthroscopy, surgery, inpatient ward stays, day case admissions, outpatient consultations, unanticipated postoperative events and their associated consultations, treatments and admissions, and investigations and interventions related to the original knee surgery, including echocardiograms, histology, X-rays, MRI scans, and ultrasound. Information was collected from the first preoperative outpatient appointment relating to original ACI or mosaicplasty knee surgery to 2 years postoperatively.

Costs for resource use were obtained from the RNOH financial services department for 2003–04 (Table 1). Costs comprised operation costs (including implant culture for ACI and instrument hire for mosaicplasty), ward costs (surgical, rehabilitation, and day case), outpatient (first and follow-up appointments), physiotherapy, and further tests and procedures. Adjustments were made to the costs when participants received several interventions concurrently. Costs for the most expensive procedure were included together with half the cost of the secondary intervention. This strategy allows for the economies of scale gained by undertaking two procedures concurrently—for example, savings in terms of administration and staff time.

Postal Questionnaire

All participants were asked about age, sex, and current occupational status as well as knee-related and general health status. Given the cross-sectional retrospective design of the study, participants were unable to complete the postal questionnaire at a consistent time point postoperatively. The main condition-related measure was the Modified Cincinnati Knee Rating System, which evaluates symptoms, overall knee condition, and activities of daily living according to pain (0–20), swelling (0–10), giving way (0–20), overall activity level (0–20), walking (2–10), stairs (2–10), running activity (1–5), and jumping or twisting activity (0–5) (2;34). Pain was evaluated using the Pain Disability Index (PDI), which evaluates interference with normal roles caused by chronic pain on a rating scale from 0 (no pain-related disability) to 10 (total disability) along seven activity categories: family/home responsibilities, recreation, social activity, occupation, sexual behavior, self-care, and life-support activities (e.g., eating, sleeping, and breathing) (12;38). Global condition-related life satisfaction was evaluated by satisfaction ratings with “living the rest of their life with their knee condition the way it is now,” on a scale ranging from 1 (delighted) to 7 (terrible) (44).

General health status was collected with the EQ-5D (9). The EQ-5D evaluates participants' health status “today” along five dimensions: mobility, ability to undertake self-cares, ability to participate in usual activities, pain/discomfort, and anxiety/depression. For each dimension, participants select one of three options—[1] no problems, [2] moderate problems, and [3] extreme problems. The focus for this study is on the social values (utility) derived for the possible 245 EQ-5D health states valued by the general UK population using the time trade-off method (14). For example, a person with a health status profile of “no problems” in any of the five dimensions (11111) is allocated a social tariff score of 1.00 (best possible health state), and a person with moderate problems in the dimensions of usual activities and pain/discomfort (11221) a social tariff score of 0.76. The EQ-5D was selected because it is simple to complete and widely used and it has been found to perform reliably and validly for a range of groups and diseases (4;9). Importantly, it is a utility measure, and no studies of ACI or mosaicplasty appear to have included measures of general health status utility.

Unanticipated Surgery, Inpatient Episodes, or Outpatient Consultations

Information about unanticipated outpatient appointments, day case and inpatient admissions, and re-operations related to the original knee surgery (apart from the standard check arthroscopy at 1 year) was collected to 2 years after the original ACI or mosaicplasty.

Analysis

Data were entered into Microsoft Excel and analyzed using the Statistical Package for Social Sciences (SPSS) 11.5 (33). Average costs were calculated for all surgery recipients for 2 years postoperatively. Cost data were skewed, and there was a relatively small sample, therefore, the nonparametric bootstrap (1) was used to calculate a confidence interval around the arithmetic mean using Stata 7.0 (40). The bootstrap procedure generates a large number of independent samples by sampling with replacement from the source data. The size of each bootstrap sample taken is the same size as the original sample. The percentile method, with 1,000 replications, was used to generate 95 percent confidence intervals (5).

Between-group differences in condition-related and general health status were tested using parametric and nonparametric tests as appropriate. Only results with p≤.05 were interpreted as having statistical significance. Postoperative ACI and mosaicplasty groups were combined for comparison with the waiting list group because of the small number of mosaicplasty participants.

Neither costs nor benefits were discounted in this study. Costs tended to occur in the first year, making discounting unnecessary, and planned costs such as physiotherapy were ascribed from their inception at the start of the treatment (i.e., year 1 rather than at the end of year 2). The study was retrospective; therefore, the exact timing of postoperative benefit accrual was unknown. Final benefit at year 1 was assumed to last until the end of year 2, and benefit was not discounted.

The EQ-5D is a measure used to derive cost per quality-adjusted life years (QALYs) where both the health status magnitude and duration of benefit are considered (15). Cost utility analysis permits the cost of treatment to be considered according to the amount of benefit afforded to the patients. Cost utility analysis is particularly valuable when treatments are being provided to improve the “quality” of patients' lives—as in the case of ACI and mosaicplasty. EQ-5D data were not available preoperatively for the ACI and mosaicplasty patients. Therefore, EQ-5D results from patients currently on ACI waiting lists were used to estimate the baseline social tariff value (14). The postoperative EQ-5D results from the ACI and mosaicplasty patients provided the postoperative social tariff values.

The incremental cost effectiveness ratio (ICER) was calculated. This analysis compares the incremental cost per unit of effect of ACI with mosaicplasty (16).

Sensitivity analysis tested assumptions surrounding the comparison of ACI and mosaicplasty. Key cost drivers were hypothesized to be the cost of ACI and mosaicplasty. Unfortunately, alternative costs for mosaicplasty were not available. A Health Technology Assessment (HTA) review of ACI cited a cost for cell culture, surgery and rehabilitation of £8,547 in 2000 prices (25). For the sensitivity analysis, this calculation was uplifted to 2003 prices (£9,366) using the hospital and community health services pay and prices inflation figures (32).

RESULTS

Participant Characteristics

Of 178 people approached to participate in the study; 95 (53 percent) consented to have their records reviewed: 53 ACI recipients, 20 mosaicplasty recipients, and 22 ACI waiting list patients. Postal questionnaires were completed by: 44 ACI, 12 mosaicplasty, and 20 waiting list patients. Table 2 presents patients' characteristics according to surgery and waiting list groups. At the time of surgery, the ACI recipients tended to be younger than the mosaicplasty recipients, although this difference was not statistically significant (t=-1.4, DF=1, p=.17). Nor was there any statistically significant age difference between the ACI and mosaicplasty groups (combined) and the waiting list group (t=0.9, DF=75, p=.36). Because of small numbers in some categories, no statistical analysis of difference was undertaken according to occupational status or surgeon-identified etiology of the knee condition (Table 2).

Resource Use and Costs

Table 3 presents the average use of resources per patient for ACI and mosaicplasty. Surgical procedures included within “operations/treatments” are ACI, mosaicplasty, and any unscheduled re-operations and treatments, including arthroscopy, MUA, knee aspiration, and a lumbar sympathetic block. Inpatient stays are presented according to stays related to the main surgical treatment and “other” stays, including cell harvest for ACI recipients and postoperative rehabilitation or unanticipated readmissions for symptoms or treatments including re-operation.

Table 3 also presents the average costs per patient according to surgical group. Adjustments were made to the costs for concurrent procedures as described previously. This applied to three patients in the ACI group who received MUA at the same time as arthroscopy, and thirteen mosaicplasty patients who had the mosaicplasty operation with concurrent arthroscopy. The cost per patient was higher for ACI than for mosaicplasty; the average difference was £2,652 (95 percent confidence interval, £1,210–£3,827). This finding was statistically significant in favor of mosaicplasty.

Unanticipated Surgery, Inpatient Stays, or Outpatient Appointments after ACI or Mosaicplasty

Twenty-six (49 percent) ACI and eight (40 percent) mosaicplasty recipients experienced unanticipated surgery, inpatient stays or outpatient appointments after ACI or mosaicplasty. Nineteen ACI and six mosaicplasty patients received an unanticipated arthroscopy. Of these cases, three ACI and two mosaicplasty patients received a second unanticipated arthroscopy. Of the nine patients in the ACI group who received additional surgery other than arthroscopy, one received a second ACI, and eight received MUA (one of these eight patients had two MUA, and one had a lumbar sympathetic block and three pain clinic injections in addition to the MUA). The additional surgical procedures in the mosaicplasty group were one ACI, one MUA, and one knee aspiration.

Main Status Outcomes

Table 4 presents results for the health status instruments according to waiting list or combined ACI and mosaicplasty surgery groups. The postoperative surgery group had higher (better) mean Cincinnati scores than the waiting list group. Grouped Cincinnati severity scores found 51 percent of postoperative surgery recipients reported good or excellent scores compared with 14 percent of waiting list participants. Although the postoperative surgery groups had lower (less) pain as measured on the PDI than the waiting list group, this finding was not statistically significant. The surgery group was more satisfied than the waiting list group with the thought of living the rest of their life with their knee condition: 33 percent of the postoperative surgery group were satisfied, compared with 9 percent of the waiting list group. The EQ-5D social tariff mean scores were statistically significantly higher (better) for the combined surgery group than the waiting list group. The only statistically significant difference in the proportions of people reporting moderate or extreme problems with the five EQ-5D dimensions was for anxiety/depression: 68 percent of the waiting list group reported anxiety or depression compared with 38 percent of the surgery group.

There were no statistically significant differences between the ACI and mosaicplasty groups for the Cincinnati Knee, Life Satisfaction, or PDI scores. The EQ-5D mean social tariff score was higher (better) for the ACI group (EQ-5D mean=0.64, n=41) than the mosaicplasty group (EQ-5D mean=0.47, n=11), but this difference did not reach statistical significance (t-test=1.3, DF=12.2, p=.22).

Cost per QALY

Using the mean preoperative EQ-5D social tariff of the waiting list patients (0.41) as an estimate for the mean preoperative social tariff of our surgery patients, the improvement in utility after surgery would be 0.23 for the ACI patients and 0.06 for the small sample of mosaicplasty patients. The average cost of immediate preoperative care, surgery, and follow-up to 2 years was £10,600 for ACI recipients and £7,948 for mosaicplasty. If the estimated improvement in EQ-5D health status occurred immediately after surgery and was stable for 2 years, then the average cost per QALY would be £23,043 for ACI and £66,233 for mosaicplasty recipients.

ICER

The ICER calculated for the ACI group relative to the mosaicplasty group was £16,349. Cost data were available for all 73 patients (n=53 ACI, n=20 mosaicplasty). Outcome data were available for 52 patients (n=41 ACI, n=11 mosaicplasty). Average EQ-5D social tariff values were imputed for missing benefit data.

Sensitivity Analysis

When the lower HTA cost of £9,366 for ACI was used, the average cost per QALY fell from £23,043 to £20,361. The ICER also decreased for ACI, to £13,694.

Summary

The average cost was higher for ACI than for mosaicplasty up to 2 years after surgery (Table 3). Fewer mosaicplasty patients (n=20) than ACI patients (n=53) participated in the study. No differences in characteristics were observed between surgery patients and people currently on a waiting list for ACI (Table 2). Postoperatively, ACI and mosaicplasty patients (combined) experienced better health status than people waiting for ACI according to the Cincinnati Knee Score, Life Satisfaction scores, PDI, and EQ-5D social tariff (Table 4). Postoperatively, ACI patients tended to have better health status outcomes than mosaicplasty patients, although differences were not statistically significant. In the 2 years after ACI or mosaicplasty, 49 percent of ACI patients and 40 percent of mosaicplasty patients experienced unanticipated consultations, admissions, and surgery.

DISCUSSION

Two other studies have considered the costs and consequences of ACI (28;30). Both studies used estimates of costs. Minas (30) estimated medical resources and consequences for 44 patients receiving ACI in Massachusetts. Minas estimated prices for direct medical resources (surgery, biopsy, cell cultures for the procedure, and care for 12 months after ACI) from the perspective of third-party payers, with the range estimated as $17,607 to $38,400 (US$ 1997). Minas also used the ordinal ranking SF-36 physical component summary score as a quality weight to construct a cost per QALY of $6,791 extrapolated to 40 years. Lindahl et al. (28) identified a cohort of fifty-seven Swedish residents who received ACI and modeled the cost estimates for 10 years before and after ACI. Lindahl et al. estimated direct medical costs (surgery, cell processing, and physiotherapy) and included an estimate for sick days. The estimated costs (US$ 1998) were $5,875 before ACI and $881 after ACI—because sick days were greatly reduced after ACI.

Neither study directly compared ACI with alternative treatments. Unlike our study, they could not address the economic argument for undertaking ACI over competing alternatives. Although our study was retrospective, direct resource use per patient was quantified, linking service delivery closely with actual resource use. Furthermore, our study used a preference-based utility measure to estimate QALYs. This strategy is preferable to using an ordinal or ranking tool (30).

A limitation of our study was the collection of secondary-care NHS costs alone. This finding reflects the reality of research involving groups of patients with complex long-term conditions and also the difficulties associated with collecting longer-term data in primary care. However, it would be helpful to funders, clinicians, and patients, if primary-care costs and direct costs incurred by patients (and their families) related to their knee condition were also known.

Future research should prospectively include general measures of utility such as the EQ-5D. In our study, health status utility was collected cross-sectionally within groups. Postoperative recovery from both ACI and mosaicplasty is not immediate, and a period of rehabilitation is necessary. The estimated EQ-5D improvement would not be immediate, and the true cost per QALY ratios would not be as favorable over a 2-year period. For example, if recovery took 6 months before improvement, the cost per QALY ratio would then be £30,725 for ACI and £88,311 for mosaicplasty. Furthermore, the estimated cost per QALY and ICER results are based on results from only 11 mosaicplasty participants. However, our results do indicate a trend toward improved health status in the postoperative ACI group compared with the mosaicplasty group. Elsewhere, other researchers have found that 98 percent of ACI recipients with good or excellent knee-related outcomes 2 years postoperatively, maintain this status over a 10-year period (36). If EQ-5D social tariff values are found to be similarly durable over the longer term, the cost per QALY ratio becomes increasingly favorable for ACI recipients. Medium-term clinical outcomes after mosaicplasty have also been encouraging, although the long-term graft durability for mosaicplasty remains unknown (19;20;23).

Direct comparison of our health status outcomes with those of other studies is inappropriate due to significant differences of study design—particularly the assessment protocol. However, our rate of repeat arthroscopy after ACI is higher than reported elsewhere. Furthermore, a considerable proportion of ACI and mosaicplasty patients came back for other unanticipated consultations and treatments. ACI and mosaicplasty are comparatively new technologies, and because the participants were part of a trial, patients were explicitly encouraged to come back if they had any concerns at all after surgery. A lower rate of graft hypertrophy would be expected if all procedures were performed using a collagen cover rather than periosteum. Future, long-term follow-up of routine postoperative ACI and mosaicplasty patients is required to determine with greater certainty the risks and benefits associated with both treatments.

POLICY IMPLICATIONS

To our knowledge, no other studies have explored the costs and utility of ACI compared with alternative current “best” treatments for people with chondral knee problems. Such research is difficult for several reasons. First, for many patients, their knee disorders are long-standing and they have already received several treatments, including surgery, before ACI or mosaicplasty. Second, because patients are participating in research investigating new technologies, they may have presented to the research team with adverse events or may have received additional investigations (such as the 1-year follow-up check arthroscopy) that would not occur if the treatments were routine. Despite such difficulties, this comparative information is necessary to inform both national, local, and individual-level doctor–patient decision making.

Although our study was retrospective, direct resource use per patient was quantified, linking service delivery closely with actual resource use and a preference-based utility measure was used to estimate QALYs. Our study has found differences in average cost between ACI (£10,600) and mosaicplasty (£7,948) favoring mosaicplasty over ACI. However, cost per QALY estimates demonstrate the importance of not considering costs in isolation from patient-experienced health status outcomes. When cost is considered alongside patient health status utility benefits, ACI (£23,043 per QALY) is more favorable than mosaicplasty (£66,233 per QALY). There is currently debate about the financial threshold for funding new health treatments within England. An implicit threshold suggests that new technologies need to cost £30,000 or less (per QALY) to obtain state health funding (43). Both the estimated cost per QALY (£23,043) and the relative ICER for providing ACI over mosaicplasty (£16,349) fell beneath this funding threshold, as did the corresponding estimates in the sensitivity analysis.

Funders, clinicians, and patients would find it helpful if primary-care costs and direct costs incurred by patients (and their families) related to their knee condition were known. However, our study has confirmed that the ACI technique results in cost-effective improvements in health status. Prospective studies should include measures of utility to confirm the estimated cost utility ratios of ACI and mosaicplasty—and to compare these with other procedures currently being provided.

CONTACT INFORMATION

Sarah Derrett, BA, DipCpN, MPH, PhD, Lecturer (), Elizabeth A. Stokes, BSc (Hons), Research Assistant (), Marilyn James, PhD, MSc, BA (Hons), Senior Lecturer (), Centre for Health Planning and Management, Darwin Building, Keele University, Staffordshire ST5 5BG, UK

William Bartlett, MRCS(Ed), BSc, MBBS, Research Fellow (), Joint Reconstruction and Cartilage Transplantation Unit, George Bentley, DSc, ChM, FMedSci, Professor of Orthopedic Surgery (), Royal National Orthopaedic Hospital, Brockley Hill, Stanmore, Middlesex HA7 4LP, UK

We thank Richard Little (Keele University) for helpful comments on an earlier version of this paper. We are most grateful to Caroline Snell for collecting the data from the records of consenting patients. We also thank Jonathan Tymms (RNOH) for providing the cost information and Chris Gooding (formerly at RNOH) for his helpful discussions with us regarding design and data collection. Funding for this study was provided to the academic and National Health Service host institutions by Verigen UK Limited.

References

Barber JA, Thompson SG. 2000 Analysis of cost data in randomised trials: An application of the non-parametric bootstrap. Stat Med. 19: 3219– 3236.Google Scholar
Barber-Westin SD, Noyes FR, McCloskey JW. 1999 Rigorous statistical reliability, validity, and responsiveness testing of the Cincinnati Knee Rating System in 350 subjects with uninjured, injured, or anterior cruciate ligament-reconstructed knees. Am J Sports Med. 27: 402– 416.Google Scholar
Bentley G, Biant LC, Carrington RWJ, et al. 2003 A prospective, randomised comparison of autologous chondrocyte implantation versus mosaicplasty for osteochondral defects in the knee. J Bone Joint Surg Br. 85: 223– 230.Google Scholar
Brazier JE, Harper R, Munro J, Walters SJ, Snaith ML. 1999 Generic and condition-specific outcome measures for people with osteoarthritis of the knee. Rheumatology (Oxford). 38: 870– 877.Google Scholar
Briggs AH, Wonderling DE, Mooney CZ. 1997 Pulling cost-effectiveness analysis up by its bootstraps: A non-parametric approach to confidence interval estimation. Health Econ. 6: 327– 340.Google Scholar
Briggs TWR, Mahroof S, David LA, et al. 2003 Histological evaluation of chondral defects after autologous chondrocyte implantation of the knee. J Bone Joint Surg Br. 85: 1077– 1083.Google Scholar
Brittberg M, Lindahl A, Nilsson A, et al. 1994 Treatment of deep cartilage defects in the knee with autologous chondrocyte transplantation. N Engl J Med. 331: 889– 895.Google Scholar
Brittberg M, Winalski CS. 2003 Evaluation of cartilage injuries and repair. J Bone Joint Surg Am. 85 (Suppl 2): 58– 69.Google Scholar
Brooks R, The EuroQol Group. 1996 EuroQol: The current state of play. Health Policy. 37: 53– 72.Google Scholar
Buckwalter JA, Mankin HJ. 1997 Articular cartilage. Part II: Degeneration and osteoarthritis repair, regeneration and transplantation. J Bone Joint Surg Am. 79: 612– 632.Google Scholar
Chang RW, Falconer J, Stulberg SD, et al. 1993 A randomised, controlled trial of arthroscopic surgery versus closed-needle joint lavage for patients with osteoarthritis of the knee. Arthritis Rheum. 36: 289– 296.Google Scholar
Chibnall JT, Tait RC. 1994 The Pain Disability Index: Factor structure and normative data. Arch Phys Med Rehabil. 75: 1082– 1086.Google Scholar
Curl WW, Krome J, Gordon ES, et al. 1997 Cartilage injuries: A review of 31,516 knee arthroscopies. Arthroscopy. 13: 456– 460.Google Scholar
Dolan P, Gudex C, Kind P, Williams A. 1995. A social tariff for EuroQol: Results from a UK general population survey. Discussion Paper 138. University of York: Centre for Health Economics;
Drummond MF, O'Brien BJ, Stoddart GL, Torrance GW. 1997. Methods for the economic evaluation of health care programmes. 2nd ed. Oxford: Oxford University Press;
Drummond MF, O'Brien BJ, Stoddart GL, Torrance GW. Methods for the economic evaluation of health care programmes. 2nd ed. Oxford: Oxford University Press; 1997 130– 131.
Friedman MJ, Berasi CC, Fox JM, et al. 1984 Preliminary results with abrasion arthroplasty in the osteoarthritic knee. Clin Orthop. 182: 200– 205.Google Scholar
Haddo O, Mahroof S, Higgs D, et al. 2004 The use of chondrogide membrane in autologous chondrocyte implantation. Knee. 11: 51– 55.Google Scholar
Hangody L, Feczkó P, Bartha L, Bodó G, Kish G. 2001 Mosaicplasty for the treatment of articular defects of the knee and ankle. Clin Orthop. 391 (Suppl): S328– S336.Google Scholar
Hangody L, Füles P. 2003 Autologous osteochondral mosaicplasty for the treatment of full-thickness defects of weight-bearing joints. J Bone Joint Surg Am. 85 (Suppl 2): 25– 32.Google Scholar
Hangody L, Kish G, Kárpáti Z, et al. 1998 Mosaicplasty for the treatment of articular cartilage defects: Application in clinical practice. Orthopedics. 21: 751– 756.Google Scholar
Henderson I, Tuy B, Oakes B. 2004 Reoperation after autologous chondrocyte implantation. J Bone Joint Surg Br. 86: 205– 211.Google Scholar
Horas U, Pelinkovic D, Herr G, Aigner T, Schnettler R. 2003 Autologous chondrocyte implantation and osteochondral cylinder transplantation in cartilage repair of the knee joint. A prospective, comparative trial. J Bone Joint Surg Am. 85: 185– 192.Google Scholar
Jackson RW, Marans HJ, Sliver RS. 1988 Arthroscopic treatment of degenerative arthritis of the knee. J Bone Joint Surg Br. 70: 332.Google Scholar
Jobanputra P, Parry D, Fry-Smith A, Burls A. 2001 Effectiveness of autologous chondrocyte transplantation for hyaline cartilage defects in knees: A rapid and systematic review. Health Technol Assess. 5: 1– 57.Google Scholar
King PJ, Bryant T, Minas T. 2002 Autologous chondrocyte implantations for chondral defects of the knee: Indications and technique. J Knee Surg. 15: 177– 184.Google Scholar
Knutsen G, Engebretsen L, Ludvigsen TC, et al. 2004 Autologous chondrocyte implantation compared with microfracture in the knee: A randomized trial. J Bone Joint Surg Am. 86: 455– 464.Google Scholar
Lindahl A, Brittberg M, Peterson L. 2001 Health economics benefits following autologous chondrocyte transplantation for patients with focal chondral lesions of the knee. Knee Surg Sports Traumatol Arthrosc. 9: 358– 363.Google Scholar
Micheli LJ, Browne JE, Erggelet C, et al. 2001 Autologous chondrocyte implantation of the knee: Multicenter experience and minimum 3-year follow-up. Clin J Sport Med. 11: 223– 228.Google Scholar
Minas T. 1998 Chondrocyte implantation in the repair of chondral lesions of the knee: Economics and quality of life. Am J Orthop. 11: 739– 744.Google Scholar
Minas T. 2001 Autologous chondrocyte implantation for focal chondral defects of the knee. Clin Orthop. 391 (Suppl): S349– S361.Google Scholar
Netten A, Curtis L. 2003 Unit costs of health and social care 2003. Canterbury: Personal Social Services Research Unit, University of Kent; 131.
Nie NH, Hull CH, Jenkins JG, Steinbrenner K, Bent DH. 1970. SPSS: Statistical package for the social sciences. New York: McGraw-Hill Book Company;
Noyes FR, Barber SD, Mooar LA. 1989 A rationale for assessing sports activity levels and limitations in knee disorders. Clin Orthop. 246: 238– 249.Google Scholar
Peterson L, Brittberg M, Kiviranta I, Akerlund EL, Lindahl A. 2002 Autologous chondrocyte transplantation. Biomechanics and long-term durability. Am J Sports Med. 30: 2– 12.Google Scholar
Peterson L, Minas T, Brittberg M, Lindahl A. 2003 Treatment of osteochondritis dissecans of the knee with autologous chondrocyte transplantation. J Bone Joint Surg Am. 85 (Suppl 2): 17– 24.Google Scholar
Pirdie AH. 1959 The method of resurfacing osteoarthritic knees. J Bone Joint Surg Br. 41: 618– 623.Google Scholar
Pollard CA. 1984 Preliminary validity study of the Pain Disability Index. Percept Mot Skills. 59: 974.Google Scholar
Sledge SL. 2001 Microfracture techniques in the treatment of osteochondral injuries. Clin Sports Med. 20: 365– 377.Google Scholar
StataCorp. 2001. Stata statistical software: Release 7.0. College Station, TX: Stata Corporation;
Steadman JR, Rodkey WG, Briggs KK. 2002 Microfracture to treat full-thickness chondral defects: Surgical technique, rehabilitation and outcomes. J Knee Surg. 15: 170– 176.Google Scholar
Tippet JW. 1991 Articular cartilage drilling and osteotomy in osteoarthritis of the knee. In: McGinty JB, ed. Operative arthroscopy. New York: Raven Press; 325– 339.
Towse A, Pritchard C, Devlin N, eds. 2002. Cost-effectiveness thresholds: Economic and ethical issues. London: Kings Fund and Office of Health Economics (UK);
World Health Organisation. 1993. Proceedings of the 2nd International Consultation on Benign Prostatic Hyperplasia (BPH). Paris: World Health Organisation;
Figure 0

Resource Unit Costs for the 2003/2004 Financial Year for Autologous Chondrocyte Implantation and Mosaicplasty-Related Secondary Care

Figure 1

Participant Characteristics at Time of Surgery or Postal Questionnaire According to Surgical Group

Figure 2

Average Resource Use and Cost per Patient According to Surgical Group over 2 Years

Figure 3

Condition-Related and General Health Status for Participants According to ACI Waiting List Group and Combined Postoperative ACI and Mosaicplasty Groups