Impact of arthroscopy of the hip for femoroacetabular impingement on quality of life at a mean follow-up of 3.2 years

Patients and Methods

This is a single-surgeon prospective series of patients with a minimum follow-up of one year who underwent arthroscopy for FAI, assessed from a QoL perspective. All the patients were asked to respond to questionnaires in order to provide a modified Harris hip score (HHS)^15,16 and their view of their satisfaction with surgery. Data on intra-operative findings and the procedure performed were meticulously recorded at the time of the operation. The scores were collected immediately pre-operatively on the day of surgery, and post-operatively at six weeks, six months and one year. All information was entered into a prospectively maintained database.

Between January 2005 and December 2009, 612 patients with FAI were operated on by the senior author (RNV). We excluded bilateral arthroscopies to reduce the impact of the contralateral side on the scores. The mean age of the patients at the time of surgery was 36.7 years (14 to 75), and there were 257 women (42%) and 355 men (58%). Surgery involved removal of the impingement lesion (femoral and/or acetabular) and labral preservation with, where appropriate, repair of chondral/labral/chondrolabral lesions. Femoral osteochondroplasty alone was performed in 537 hips (88%), combined femoral osteochondroplasty with acetabular recession was performed in 61 hips (10%) and acetabular recession alone in 14 hips (2%). Chondral and labral lesions were treated using thermal treatment with radiofrequency probe, labral reattachment, or chondrolabral repair using either sutures or fibrin adhesive. Thermal treatment alone was performed in 469 hips (77%), labral or chondrolabral repair with suture anchors (Bioraptor; Smith & Nephew UK Ltd, London, United Kingdom) or the Fast-Fix meniscal repair system (Smith & Nephew UK Ltd) was performed in 78 hips (13%), and fibrin glue was used in 65 hips (10%). All patients had FAI and a labral tear, with a wide range of associated pathologies (osteoarthritis in 137 hips, osteochondral lesions in 132, chondral delamination in 66, psoas tendonitis in 39, ligamentum teres tear in 51, loose bodies in 23 and trochanteric bursitis in eight). Post-operatively all patients undertook a standardised regimen. Briefly, for the first four weeks patients were asked to be partially weight-bearing with crutches. Formal physiotherapy was started one week after surgery and continued for four months, using a documented rehabilitation plan.

We used the answers from the modified HHS to translate it to a Rosser Index Matrix-created QoL score¹⁷ using previously reported techniques.^18-20 The qualitative nature of the questions from the modified HHS^18-20 allows them to be translated to Rosser distress (pain) and disability (function) categories (Table I), which can be applied to the Rosser Index Matrix²¹ (Table II) to derive QoL scores. The Rosser Index Matrix allocates scores from -1.486 to 1.000, and there are 29 possible health states, each with a numerical score for QoL.²¹ A score of 1.000 indicates normality, whereas death is given a score of 0.000. Negative scores indicate health states thought to be worse than death. The patients were also asked whether they were satisfied with the results of the surgery, and to respond ‘yes’ or ‘no’.

Results

The mean follow-up was 3.2 years (1 to 7).

Outcome for the whole group

The mean QoL score increased from 0.946 (-1.486 to 0.995) pre-operatively to 0.974 (0.7 to 1) at one year after surgery (p < 0.001, t-test) (Fig. 1).

Fig. 1

Plot for the whole group depicting the pre-operative and one-year post-operative quality of life (QoL) scores. The error bars show the 95% confidence intervals.

At one-year follow-up the QoL score showed at least a one-grade improvement in 469 patients (76.7%), no change in 88 (14.4%), and in 55 (8.9%) the score deteriorated.

At one year after surgery 450 patients (73.5%) were happy with the results. All the patients who were pleased with the results of the surgery had an improvement in QoL score. In all, 19 patients were not pleased with the surgery despite an improvement in QoL score, either because it did not match their expectations or because of the prolonged rehabilitation involved.

Influence of gender difference

The mean QoL score for men was significantly better than for women, both before surgery (0.962 (0.7 to 0.99) versus 0.924 (0 to 0.99); p < 0.001, t-test) and at one year (0.982 (0.7 to 1) versus 0.964 (0.7 to 1); p < 0.001, t-test) after surgery (Fig. 2). However, the mean change in the QoL score was greater for women (0.02 (-0.21 to 0.27) for men, 0.04 (-0.16 to 0.87) for women), although this was not statistically significant (p = 0.12, t-test) (Fig. 3).

Fig. 2

Plot showing the gender difference between the pre-operative (Pre-op) and one-year post-operative (Post-op) quality of life (QoL) scores. The error bars show the 95% confidence intervals.

Fig. 3

Plot showing the gender difference in the change in quality of life (QoL) scores at one year after surgery. The error bars show the 95% confidence intervals.

Influence of difference in age

The groups were split into those patients ≤ 50 years of age (n = 540) and those > 50 (n = 72). The mean QoL score for the two groups (Fig. 4) did not show any significant difference either before surgery (≤ 50 years: 0.944 (0.7 to 0.99); > 50 years: 0.952 (0.7 to 0.986); p = 0.67, Mann-Whitney U test) or at one year after surgery (≤ 50 years: 0.974 (0.7 to 1); > 50 years: 0.967 (0.7 to 1); p = 0.37, Mann-Whitney U test).

Fig. 4

Plot showing the influence of difference in age between the pre-operative (Pre-op) and one-year post-operative (Post-op) quality of life (QoL) scores. The error bars show the 95% confidence intervals.

The mean change in QoL score (Fig. 5) was greater for the patients ≤ 50 years of age (≤ 50 years: 0.04 (-0.21 to 0.87); > 50 years: 0.01 (-0.1 to 0.13)), but the difference was not statistically significant (p = 0.39).

Fig. 5

Plot showing the influence of difference in age (in years) on the change in quality of life (QoL) score at one year after surgery. The error bars show the 95% confidence intervals.

Discussion

In a large prospective series we found that QoL improves after hip arthroscopy, labral preservation, femoral osteochondroplasty and acetabuloplasty in patients with FAI. The QoL scores improved in 469 patients (76.7%) and 450 (73.5%) were happy with the outcome of surgery at one year. We noted that the QoL scores were significantly (p <  0.001) lower in women, both pre- and post-operatively. A linear regression analysis revealed that the best predictor of change in QoL score is the pre-operative QoL score (p < 0.001) followed by the gender of the patient (p = 0.04).

We generated the QoL scores from a patient-derived classification of health states called the Rosser Index Matrix.¹⁷ The index is the most commonly used in United Kingdom, making the results of this study comparable with studies that look at QoL outcome for other interventions.^22-24 The translation of pre-existing scores to the Rosser Index Matrix can be a source of error and subjectivity.²⁴ However, the modified HHS is closely comparable to the Rosser categories, allowing the two parameters (pain and function) to be translated and thereby quantified to a QoL score.^18-20,25 Whereas the modified HHS has values ranging from 0 to 100, the change in QoL score will fall into one of 29 categories (26 different scores, after excluding some common scores). Therefore, a minor change in the modified HHS does not equate to a change in QoL, whereas a major change does. Any change in QoL can therefore be considered to be more clinically relevant in the overall health status of an individual.

The age of the patient and the presence of associated diagnoses such as osteoarthritis, osteochondral defects, chondral delamination, ligamentum teres tear, psoas tendonitis, loose bodies and trochanteric bursitis did not have a bearing on the QoL score. Those patients over the age of 50 had statistically similar scores to patients under the age of 50 (p = 0.39), although the mean change in score was lower for the former, older group (Fig. 5). Others have also reported the irrelevance of older age when predicting outcome after arthroscopic surgery for FAI. Javed and O’Donnell²⁶ reported the results of arthroscopic femoral osteochondroplasty for cam FAI in patients over the age of 60 and found an overall satisfaction rate of 75%.

In our study, the pre- and post-operative QoL scores were significantly lower for women (p < 0.001) than for men. Although the change in score was higher for women, this was not statistically different (p = 0.12). We could not find such a difference in outcome in previous reports. We established by linear regression analysis that the gender of the patient was a significant predictor (p = 0.04) of a change in QoL score. However, the best predictor of change in QoL was the pre-operative QoL score: the lower the pre-operative QoL score, the greater the change in QoL after surgery (ρ = -0.66, p < 0.001). This information is crucial for the purpose of informed consent.

When considering our whole study group, at one year after surgery the QoL scores improved in 76.7% of patients, remained unchanged in 14.4%, but deteriorated in 8.9%. This is similar to other studies in the general population, which have revealed that arthroscopic management of FAI leads to a good-to-excellent result in up to 75% of patients at one year.^27,28 Even in patients with milder degrees of pre-operative degenerative change on their radiographs, osteochondroplasty for FAI can lead to an improvement in pain and function at short-term follow-up.²⁹ In a series of 200 patients, Byrd and Jones³⁰ reported that 83% of patients had an improvement in their HHS at a mean follow-up of 16 months. In a series of 34 patients with FAI treated by open surgical dislocation, labral debridement and femoral osteochondroplasty, Beaulé, Le Duff and Zaragoza³¹ demonstrated that QoL, in terms of Short Form-12 scores, improved after surgery in 82% of patients. However, studies on the arthroscopic management of FAI have not measured the improvement in terms of change in QoL, which is essential for complete assessment of the benefit of an intervention.³

This study has a selection bias owing to the nature of the tertiary referral pattern. However, it is nevertheless a prospective study of a large cohort of patients operated upon by a single surgeon. We looked uniquely at the impact of surgery on QoL, which has not been explored previously, and found that arthroscopic surgery for FAI improves QoL in 76.7% of patients. The pre-operative QoL score and gender were significant predictors of the change in QoL.