Mid-term outcomes of arthroscopic-assisted Core decompression of Precollapse osteonecrosis of femoral head—minimum of 5 year follow-up

Osteonecrosis of the femoral head (ONFH) is a progressive disease caused by an interruption of blood flow to the femoral head that leads to collapse and the development of secondary arthritis. Mont et al. estimated the prevalence of total hip arthroplasty (THA) involving osteonecrosis to be between 8—12%. With an estimated annual volume of THA of 371,000, it is projected that roughly 37,000 cases of THA involve ONFH [1‐3]. Epidemiologically, patients in the 30’s to 40’s are commonly affected, resulting in a significant functional impairment in a relatively young population culminating in a loss of productivity for both the individual and society [4]. There are numerous direct and indirect risk factors associated with the development of ONFH. Direct factors include fractures, dislocation, sickle cell disease, and radiation. Indirect factors include corticosteroid use, excessive alcohol intake, certain metabolic syndromes, and smoking [3, 5]. Overall, the most commonly implicated risk factors are corticosteroids, alcohol, and trauma. Finally, it is also common for cases to occur idiopathically with no identified risk factor [6‐8].

The preferred management of ONFH remains controversial and is largely dependent on the staging of the lesion. While there are several classification systems, the Ficat-Arlet staging is commonly used, with stages 0 through IIb considered precollapse, and stages III and IV considered collapsed [3, 9]. (Table 1) Management options can be categorized into observation, non-operative treatment, and operative treatment. Observation may allow small asymptomatic lesions to spontaneously resolve, although most lesions will progress [10]. Lieberman et al. found that 67% of asymptomatic hips and 85% of symptomatic hips managed with observation progress to collapse [11]. Non-operative therapies include bisphosphates, statins, prostaglandins, and hyperbaric oxygen. These treatments have been shown to decrease pain, but long-term outcomes are uncertain and require randomized control trials [10, 12]. Mont et al. literature review proposes that non-operative treatment may be attempted in asymptomatic hips, but once symptoms manifest operative treatment is indicated [7].

Operative management is divided into two main categories: femoral head-preserving and arthroplasty [13]. Femoral head-preserving techniques have demonstrated successful outcomes in precollapsed staged lesions. Within the femoral head-preservation category there are numerous methods, variations, and adjuvants. The treatment algorithm for determining which method to pursue is largely based on staging and is still being delineated. Common head-preserving methods includes: 1) standard core decompression (CD), 2) multiple drilling decompression, 3) femoral rotational osteotomies, 4) non-vascularized bone graft, 5) vascularized bone graft, and 6) tantalum rod placement. Once collapse has occurred arthroplasty is indicated with excellent outcomes. Arthroplasty consists of either resurfacing or hip replacement [10, 12]. Ancelin et al. found similar THA survivability rates in patients with ONFH and primary osteoarthritis at ten-year follow-up [14].

Although there is no standard approach to the management of ONFH, CD is considered the most common treatment for early stage lesions. This management is based on the principle of decreasing intraosseous hypertension that is created by the necrosis process and inflammatory cell infiltration into the affected areas [10]. CD is performed percutaneously, using an 8—10 mm trephine that targets the area of necrosis using fluoroscopic guidance. Whereas, multiple drilling using a smaller diameter trephine to create more than one core track, and has achieved favorable outcomes while having lower complication rates, including subtrochanteric fracture [13, 15]. Al Omran et al. compared standard core decompression and multiple drilling in a cohort of patients with sickle cell disease, finding no statistically significance in outcomes or complications, although the multiple drilling technique had the benefit of being less invasive [16].

Although the literature has considered the various surgical techniques, arthroscopic management of ONFH is a new and evolving approach for hip preservation. One of the earliest descriptions of arthroscopic management of ONFH was reported by Ruch et al. in which arthroscopy was used to document accurate positioning of the core decompression [17]. Since Ruch’s paper in 1998, arthroscopic management has developed a relatively limited presence in the literature: one review article by Papavasiliou et al., several technique papers, and very few cohort analyses [4, 18‐23]. With this limited body of literature, the purpose of this study is to provide further evaluation of arthroscopic-assisted core depression in the treatment of patients with ONFH, while being the first study to attain mid-term follow-up. We hypothesize that this cohort analysis will demonstrate that hip arthroscopy is able to treat ONFH by achieving accurate and minimally invasive decompression; in addition to successfully addressing concomitant intraarticular pathologies resulting in reliable mid-term survivability and an acceptable rate of conversion to THA.

A total cohort of 11 hips (8 patients) was identified, of which 1 patient (12.5%) was female and 7 patients (87.5%) were male. All eight patients were included based on the aforementioned inclusion and exclusion criteria. The mean age of the cohort was 36.4 years ±9.2 years (range, 17–48). (Table 2) The mean center-edge angle was 30.5° ± 3. 4° (range 26°— 36°). Patients’ past medical history was evaluated for established risk factors for the development to ONFH. Two hips (18.2%) had inhaled corticosteroid exposure and 4 hips (36.4%) had a history of opioid abuse. (Table 3).

The Ficat-Alret classification found on preoperative imaging was Stage I in 3 (27.2%), IIa in 4 (36.4%), and IIb in 4 (36.4%) hips. (Table 4) The mean preoperative symptom duration from onset of symptoms to arthroscopic intervention was 14.5 months ±9.04 months (range, 4—36 months). Four (36.4%) hips experienced mechanical issues, including locking, catching, or buckling. The most common concomitant pathology addressed at the time of arthroscopy, was labral repair/debridement, followed by microfracture. (Table 4) The mean traction time was 56.7 min ± 11.4 min (range, 43—78) and mean operative time was 100.9 min ± 16.6 min (range, 85—130). Arthroscopic verification of guide pin and reamer placement was achieved in all 11 surgeries. Finally, there were no major complications (0%), including sub-trochanteric fracture or violation of the articular cartilage; or minor complications.

The primary outcome of this study was conversion to THA. The mean patient follow-up was 7 years ±1.48 years (range, 64—118 months). At final follow-up, 6 hips (54.5%) had not converted to THA. Upon further stratification, Stage I—100%, Stage IIa—75%, for a combined 87%, had not converted to THA, in contrast, 100% of hips categorized as Stage IIb had converted to THA. (Table 5) The indication for all five conversions to THA was symptomatic and radiographic progression of the ONFH. Of the five hips that converted to THA, the time interval from hip arthroscopy to THA was 22.6 months ±9.5 months (range, 7—33 months).

Conversion to THA was analyzed for correlation with other variables and was found to be associated with Ficat-Alret staging of the hip. (p-value = 0.03) (Table 5) Further analysis found that this correlation was particularly significant for the Stage IIb subset of hips. (p-value = 0.015) The correlation of conversion to THA with mechanical symptoms, inhaled corticosteroid exposure, smoking tobacco exposure, and bilateral hip involvement of ONFH, was not statistically significant.

This study demonstrated good survivability of arthroscopic-assisted standard core decompression (CD) in precollapse ONFH with no complications at mid-term follow-up. In this cohort of 11 hips with an average follow-up of 7 years (with a minimum of 5 years), 6 patients (54.5%) had not converted to THA. (Fig. 3) However, when stratified by lesion stage, 100% of Stage I hips, 75% of Stage IIa, and 0% of Stage IIb hips had not converted to THA.

This study’s survivability is consistent with recently published data. Mukisi-Makaza et al. reported survivability in sickle cell patients with an average follow-up of 7.4 years utilizing the Ficat classification: 100% for Stage I and 47% for Stage IIa (did not report the conversion rate in Stage IIb lesions) [25]. Yang et al. reported similar rates using the Steinberg classification: 61% for Stage II and 25% in Stage III [26]. Another study with shorter follow-up of 3 years, evaluated CD with physical therapy reported similar rates for early lesions: 90% for Stage I and 82% for Stage II [27]. Finally, utilizing multiple drilling, Al Omran et al. reported similar survivability rates at a mean follow-up of 6.1 years: 83% for Stage I, 79% for Stage IIa, and 69% for Stage IIb [28].

Several studies have found that CD is more effective in the treatment of early avascular necrosis in the precollapse stage, particularly when the lesion is small (Keboul Necrotic Angle < 160°) or medium-sized (Keboul Necrotic Angle between 161—199°) [7]. This is congruent with this study finding high survivability for Stages I and IIa, but poor survivability for Stage IIb. Although this study has a small sample size, these findings indicate that arthroscopic-assisted management should be focused on earlier staged ONFH.

In this cohort, there were no (0%) major or minor complications, including femoral fracture or violation of the cartilage. Comparatively, percutaneous standard CD is associated with a 10—15% complication rate, including weakening the bone contributing to collapse, subtrochanteric fracture, or violation of the articular cartilage [7, 12]. Therefore, arthroscopic-assisted CD is associated with fewer complications [21].

Several variables were evaluated for correlation with conversion to THA. Ficat-Alret staging was found have a statistically significant association, particularly for Stage IIb lesions. This makes sense intuitively, as advanced lesions are likely to continue to progress to collapse. This may also indicate a decreased efficacy of treating advanced ONFH with arthroscopic-assisted decompression. Interestingly, ONFH of bilateral hip involvement, mechanical symptoms, inhaled corticosteroid and smoking tobacco exposure, where not significantly associated with conversion to THA. Although, one might expect mechanical symptoms to be associated with advanced ONFH and therefore more likely to be associated with conversion to THA, these symptoms can be due to numerous intra- and extraarticular etiologies [29]. The lack of statistical significant may be due to the small sample size of this study. Similarly, Bozic et al. reported advanced radiographic stage as a significant predictor of CD failure (defined as radiographic progression to Stage III or requiring a subsequent operation). In addition, that study reported shorter duration of symptoms and use of corticosteroids as significant predictors of failure, but no association with age, gender, alcohol intake, or renal transplantation [30].

Arthroscopic management of ONFH is advantageous because it is both diagnostic and therapeutic. Guadilla et al.’s report of the arthroscopic management of ONFH by CD with adjuvant platelet-rich plasma (PRP) described the advantages of arthroscopy as: diagnosis of concomitant pathologies, precise localization and drilling of necrotic lesions, and increased avoidance of joint penetration [18]. Diagnostically, arthroscopy allows for the precise assessment of the articular surfaces of both the femoral head and the acetabulum for accurate staging and evaluation. (Fig. 4) In this study, 6 hips (54.5%) had focal changes in the articular surface of the femoral head, while 2 hips (18.2%) had diffuse changes, with parallel results seen on the acetabular side.

Unlike percutaneous standard CD, arthroscopic-assisted CD allows for the arthroscopic treatment of concomitant intra-articular pathology associated with ONFH, including: loose bodies, pincer lesions, cam lesions, synovitis, labral tears, and chondral defects [20]. (Fig. 5) In our cohort, concomitant pathologies addressed includes: labral repair/debridement in 8 (73%), microfracture in 7 (64%), femoral osteochondroplasty in 1 (9%), and synovectomy in 1 (9%), hip(s), respectively. At latest follow-up, the concomitant pathologies had resolved with the 6 hips that did not convert to THA reporting little to no pain and unremarkable clinical exam.

Of the 11 hips, 4 (36.4%) had mechanical symptoms, all of which resolved after arthroscopic treatment. (Table 4) Even with lesions in the pre-collapse stage, some patients develop mechanical symptoms such as locking, buckling, catching, popping, or clicking. In this subset of patients, the mechanical issues warrant arthroscopic treatment. Because similar to the treatment of intra-articular pathologies, these mechanical issues are not addressed by percutaneous standard CD. McCarthy et al. performed hip arthroscopy on seven patients between 1993 and 2000 with documented or radiographically confirmed avascular necrosis [31]. This study concluded that hip arthroscopy is minimally invasive and an effective treatment for early ONFH with mechanical symptoms.

Interestingly, the effect of irrigation pressure, traction, and osteoplasty terminal circulation of the femoral head is not known. Theoretically, traction or irrigation pressure could compress the terminal circulation of the femoral head, resulting in worsening of the underlying pathology of ONFH. And osteoplasty of the femoral neck in the treatment of CAM lesions could expose the trabecular bone to the pressured irrigation fluid further compressing terminal circulation. However, our practice is to utilize intermittent traction only when working in the central compartment and to use minimal irrigation pressure (pressure controlled at 40 mmHg). Furthermore, the traction and irrigation pressure are temporary factors (mean traction time of 57 min and mean operative time of 101 min), whereas CD would decrease intraosseous pressure for the future of the native hip.

Although both CD and multiple drilling decompression are used to treat ONFH, this study chose to utilize CD because it is the most common and fundamental femoral-head preserving modality. It also has the benefits of technical ease and efficiency. Unfortunately, regardless of technique, even when femoral head-preserving surgery is attempted in the early stages of ONFH a substantial portion of patients continue to progress to THA, with variable levels and durations of survivability [7]. In this study with an average follow-up of 7 years, 13% of lesions in Stage I or IIa converted to THA, while a 100% of lesions in Stage IIb converted to THA. This supports the treatment principles that head-preservation techniques achieve the best results in smaller, earlier, precollapse lesions, while larger, more advanced, collapsed lesions may necessitate THA [7, 10, 12].

Augmentation procedures of the CD and multiple drilling decompression with bone marrow grafting and calcium sulfate/phosphate injectable bioceramics are showing promise [32, 33]. Gangji et al. conducted a randomized control trial comparing CD (n = 11) versus CD with autologous bone marrow implantation (n = 13), with a follow-up of 60 months, finding a significant decrease in the progression to collapse and a reduction in pain [34]. Civinini et al. used autologous bone marrow implantation along with a bioceramic consisting of calcium sulphate and calcium phosphate to backfill the core tract from the CD (n = 37), finding only a 3.3% conversion to THA in precollapse lesions (8.1% overall) at a mean follow-up of 20.6 months [35]. However, cost, lengthening of operative duration, technical feasible, and longer term outcomes should be considered for these procedures.

There are several important limitations of this study. Despite longer follow-up compared to similar existing studies, this cohort analysis has a small sample size and further investigation with larger cohorts is necessary. Furthermore, there was no comparison group because all patients that presented with ONFH were treated similarly and there was no cohort of non-arthroscopically assisted core decompression. Finally, there are no patient-reported outcome measurements available. However, we believe that this study is contributing to the literature by providing the first report of mid-term follow-up in arthroscopically managed ONFH patients.

In conclusion, this is the longest reported follow-up of arthroscopic-assisted management of ONFH with mid-term outcomes of an average of 7 years. Hip arthroscopy-assisted core decompression is a new, evolving, and promising surgical approach to address osteonecrosis with high survivability for Stage I (100%) and Stage IIa (75%) lesions, while a 100% of Stage IIb lesions progressed to THA. In addition, hip arthroscopy allows for the treatment of the entire hip joint by addressing concomitant intraarticular pathologies and alleviating mechanical symptoms. Further studies with long-term follow-up and larger sample sizes are needed to evaluate the role of arthroscopy in ONFH.