Results
The results of hip arthroscopy are only now beginning to be mapped out. The problem has been the few centres specialising in the technique with the correspondingly small number of hip arthroscopies performed. Even with the Cambridge experience (now over 1000 cases) it is still difficult to be precise about outcomes. These cases are split further into many different categories, some of which contain too small numbers to prove statistical significance in any direction. Controlled trials are obviously needed, and yet few such studies exist in the literature for any form of arthroscopy. How many studies have been undertaken, for example, comparing arthroscopic to open meniscectomy in the knee? Yet look at how many knee arthroscopies are performed worldwide per annum.
Patients should be advised that no conclusion about the result should be drawn until 3 months following surgery. Though most recovery will have occurred within the first few postoperative weeks, the 3-month point is when symptoms will have largely stabilised.
The results for hip arthroscopy can be considered under two broad headings:
- Diagnostic
- Therapeutic
Diagnostic
It is not possible to say a hip joint is normal without looking inside the joint. As a rough guide, if a clinician is persuaded that pain is indeed emanating from the hip joint, there is an 80% chance that hip arthroscopy will find something abnormal, despite all preoperative investigations being normal.
A study has been performed between MRI - not MRI arthrography - and arthroscopy (J Bone Joint Surg 1995;77B(3):374-376). MRI was found to be unreliable for labral tears, chondral lesions less than 1 square centimetre in diameter, and loose bodies. I can recall removing over 150 loose bodies from a patient when all preoperative investigations were normal!
The difficulty facing the arthroscopic surgeon is deciding whether or not the lesion seen is genuinely responsible for the patient's problem. What is normal and what is abnormal? Would a particular lesion actually be seen in normal people without hip pain? In the early days of hip arthroscopy the stellate crease was diagnosed as an osteochondral defect, and sometimes debrided. Unsurprisingly the patients did not improve. It is therefore essential that any surgeon undertaking hip arthroscopy has a thorough understanding of the arthroscopic anatomy of the joint. Without such knowledge, overdiagnosis of suspicious lesions is a genuine risk.
Therapeutic
Results will be considered by condition encountered:
Torn acetabular labrum
Significant improvement can be expected after partial labrectomy in the absence of any other pathology within the joint. Should chondral damage be present (eg. chondral defect, osteoarthritis) results are less encouraging.
Torn ligamentum teres
Complete (Type 1) tears and degenerate (Type 3) tears do not do so well after hip arthroscopic debridement. Partial (Type 2) tears, with excision of the damaged area, do better.
Loose bodies
Removal of the multiple loose bodies found in synovial chondromatosis normally gives an excellent result. Synovial osteochondromatosis is not so predictable, once the chondral fragments have begun to ossify. Normally they are associated with some chondral damage which, in turn, limits the improvement possible. Nevertheless, some symptomatic improvement is usually seen, though it may be necessary to repeat arthroscopic debridement from time to time.
Chondral/osteochondral defects
Symptomatic improvement following hip arthroscopic debridement is usually seen, though complete resolution is unlikely. Frequently, the smaller defects are missed by MRI scanning anyway, but if detectable, healing of well demarcated defects can be seen after drilling. Healing of a defect does not always mean symptom freedom!
Synovitis
When associated with chondral destruction, symptom resolution is unlikely. Indeed a synovectomy may be technically impossible due to poor distraction. Milder forms of synovitis, eg. early PVNS (pigmented villonodular synovitis), early rheumatoid, respond well to arthroscopic synovectomy, though whether the procedure alters the long term progress of the condition is unknown.
Osteoarthritis
A variable result can be seen following hip arthroscopic debridement. For those with mild to moderate disease, with a good range of hip movement, and aged less than 46 years, there is a 70% chance of a good result two years after surgery. For patients outside this category the results are less predictable with approximately 5% being symptomatically worse three months after hip arthroscopy.
Sepsis
Excellent results are normally seen with immediate resolution of infection, even if diagnosis has been grossly delayed (5 months in one case for the Cambridge series). What arthroscopy cannot do is to recover articular cartilage damage, so a return to complete normality is only possible if sepsis has been diagnosed early, preferably within hours of onset.
Fractures and dislocations
Hip arthroscopy is unlikely to improve a patient symptomatically here. The logic of performing it is to gain a better understanding of the nature of the injury and to remove any intra-articular fragments. Whether this will allow a better prognosis, only time will tell.
Reiter's disease
The Cambridge series only includes a handful of such cases. All have improved following surgery, though this improvement has largely been in movement rather than pain levels.
Arthrofibrosis
Range of motion is usually improved, though not fully, assuming arthroscopic access has been possible.
Chondrolysis
Again, range of motion can be improved, though chondral recovery is not seen.
Chondromalacia
The so-called 'chondromalacia coxae' is a clear entity, usually affecting the same category of patient as can develop chondromalacia patellae. Arthroscopic debridement, if practical, has improved symptoms. Why improvement has occurred is a different issue!
Metabolic bone disease
Hip arthroscopy is useful for diagnosis, though will not alter the underlying condition. Symptom improvement is possible, albeit for a limited period.
Avascular necrosis
Symptom improvement following hip arthroscopy is actually possible, for reasons that are unclear. However, there is no evidence that arthroscopy will alter the long term outcome of the condition. It may, however, alter the surgical approach to its management.
N.B. Remember to tell all patients three things before hip arthroscopy:
- Arthroscopic access may not be possible
- Maximum symptomatic improvement will take three months
- There is a 5% chance of making symptoms worse
Remember also that all women of reproductive age should be asked for the date of their last menstrual period!!