Thirty patients with one-stage hip arthroplasty and multiple drilling decompression for bilateral osteonecrosis of the femoral head:outcomes at two-year follow up

Background: To investigate the postoperative outcome of decompresed hip following contralateral replacement in patients with bilateral osteonecrosis of the femoral head (ONFH). M ethods: This study retrospectively reviewed 30 patients with bilateral ONFH who underwent one-stage total hip arthroplasty (THA) and multiple drilling decompression from February 2014 to February 2016. For all patients, alendronate was prescribed. Postoperative Harris Hip Scores (HHSs), Oxford Hip Scores (OHSs), Self-Administered Patient Satisfaction (SAPS), hip internal and external rotation angles, and the time of one-leg standing were evaluated. All patients were followed up for an average of 30.33 months (ranged from 8 months to 48 months). Results: Twenty-four patients had no progression or collapse on radiographic evaluation within 2 years postoperatively. The following scores of the decompressed sides were significantly lower than those of the replaced sides: HHSs (mean: 96.5 versus 98.25, P <0.05), SAPS (mean: 96.35 versus 99.48, P <0.001), internal rotation (mean: 27.58° versus 30.50°, P <0.05), and the time of single-leg standing (mean: 24.17 s versus 31.83 s, P <0.05). There was no significant difference in OHSs and external rotation between the sides. The remaining 6 patients underwent two-stage THA on the decompressed sides within 2 years. Four of them underwent THA for severe pain and femoral head collapse. The other 2 patients showed no progression on radiographic evaluation but complained of an uncomfortable feeling in the joint. Conclusions: One-stage THA and multiple drilling decompression are effective surgical procedures for bilateral ONFH at different stages. However, a few patients with no progression on radiographic evaluation may require two-stage THA on the decompressed sides because of more pain, poorer internal rotation, and weaker gluteal muscle strength when compared with the replaced sides.

Osteonecrosis of the femoral head (ONFH) is a devastating disease that has become an increasing worldwide health problem. Although multiple theories have been proposed, no one pathophysiologic mechanism has been identified as the etiology for the development of ONFH, but the basic mechanism involves impaired circulation within a specific area that ultimately becomes necrotic [1]. The natural history of untreated asymptomatic ONFH has a high prevalence of progression to symptomatic disease and femoral head collapse [2].
Various nonoperative treatment regimens and joint-preserving procedures are indicated in the treatment of precollapse disease: nonoperative treatment includes oral medication and improved diet, physical therapy, and various combinations of non-surgical treatments; joint-preserving procedures include core decompression, femoral osteotomy, nonvascularized or vascularized bone-grafting, and stem cells [1]. Among them, core decompression of the hip is one of the most commonly performed surgical procedures that can successfully alleviate the progress of the disease and, at least, delay the patient's need for total hip arthroplasty (THA) [3]. Multiple drilling decompression is safer and less invasive than single coring, and there is no statistically significant difference in outcome or the complication rate between both procedures [4]. Once femoral head collapse is present, THA is the only effective treatment that can provide excellent outcomes at greater than 10 years of follow-up [1].
Most patients with nontraumatic ONFH have bilateral involvement [5]. For patients with bilateral involvement who have a collapsed femoral head on one side and precollapsed on the other side, one-stage THA is usually performed for the collapsed side, and jointpreserving surgery is performed for the uncollapsed side; this strategy has been proven to be a cost-effective method that does not increase perioperative morbidity [6]. Recently, Zeng et al reported good short-term to medium-term outcomes with one-sided hip-4 preserving and concurrent contralateral THA for the treatment of bilateral ONFH in different stages. Therefore, they suggested that the uncollapsed hip can achieve biological stability and sufficient blood supply through hip-preserving surgery and obtain long-term repair of the necrotic bone as well as early non-weight-bearing function training, which is possible since the whole body weight load is distributed to the hip that underwent one-stage THA [7]. However, in Zeng's study, the hip-preserving surgery was core decompression followed by tight impaction bone grafting combined with nonvascularized fibular allografting, which was a more complicated and costly procedure than multiple drilling decompression.
It is unknown whether bilateral ONFH patients with different stages can benefit from onestage THA and multiple drilling decompression and how the replaced side affects the decompressed side. Therefore, this study aimed to review a series of patients with bilateral ONFH who received one-stage THA and multiple drilling decompression, and we 200° to 240° received multiple drilling decompression [8]. The contralateral hips with 5 preoperative diagnoses of ARCO stages III or IV underwent THA. All the decompressed hip joints were treated with a 3-mm Kirschner wire for decompression of the porous core [9].
All the hips on the replacement side received uncemented hip prostheses, all of which were primary replacements performed via the posterior lateral approach. Subsequently, the patient's position was changed to supine position for decompression of the contralateral hip. All procedures were performed by the same senior surgeon. For all patients, alendronate was prescribed, and they were treated with the same perioperative management.
The survival rates of patients with decompressed sides within 2 years were measured based on both radiographic progression and using two-stage THA as the endpoint. Data between the two sides included HHSs, OHSs, and SAPS. Lower limb mobility and the standing time on a single foot were compared.

Statistical methods
All data were analyzed statistically. The Kolmogorov-Smirnov test was used to evaluate the normality of the groups' data distribution, and the Levene test was used to assess the homogeneity of variance. Paired sample t-tests were applied to normally distributed values. Data with a non-Gaussian distribution were analyzed by using the Wilcoxon signedrank test. Kaplan-Meier survivorship analyses was used to assess the Two-year survival of decompressed sides. The level of significance was set at P<0.05. All statistical analyses were performed using SPSS Statistics 19 (IBM Corp., Armonk, NY, USA).

Results
Among the 30 patients, postoperative evaluations showed that the replacement hip prosthesis was well-positioned and demonstrated good bone ingrowth, with no case of postoperative dislocation, infection, periprosthetic fracture, deep vein thrombosis, and other complications. On the decompressed side, the femoral head did not progress or 6 collapse and was well repaired in 24 patients. If THA was used as the endpoint on the decompression side, the survival rate of decompressed sides within 2 years was 80% in these one-stage THA and multiple drilling decompression Figure 1 . If the progression of necrosis or severe pain was used as the endpoint, the success rate of decompression surgery in our study would be 86.7% Figure 2 . The typical case images are shown in

Discussion
There are three clinical endpoints for decompression surgery: severe pain, radiographic progression to collapse, and THA [10]. If THA is used as the endpoint on the decompression side, the survival rate of multiple drilling decompression was 80% ( Figure   1), which was similar to the results of previous studies [11,12]. Therefore, it can be suggested that one-stage THA and multiple drilling decompression were effective for patients with bilateral ONFH at different stages. And decompressed side can obtain early non-weight-bearing function training, which benefits from distributing the whole body weight load to the hip of one-stage THA [7]. If the progression of necrosis or severe pain is used as the endpoint, the success rate of decompression surgery in our study would be 86.7% (Figure 2). Two of the 6 patients who underwent THA re-operation had no significant progression of joint disease in the decompressed hip and no severe pain.
However, the patients felt discomfort, slight pain on the decompression side, or had function that was not as good as that of the hip joint on the replacement side. This is a problem should be mentioned. This may become an important factor affecting the success rate of head preservation surgery.
We compared HHS and OHS on the decompressed and replaced sides, and found no significant difference in OHS between the two sides. However, for HHS, the difference between the two sides was statistically significant. We analyzed HHS according to the 8 module analysis and found that the pain module score of the decompressed side was lower than that of the replaced side; however, the functional module score was not significantly different. These results indicated that most patients still had slight pain after decompression. Liu et al showed that patients who underwent femoral head decompression had significantly improved pain symptoms; the postoperative visual analog scale score was about 3 points in their study, which was considered mild pain [13], and this result is similar to ours. The main causes of pain before femoral head collapse are intraosseous pressure [14], bone marrow edema, accumulation of microfractures, disruption of mechanical loading, articular cartilage injuries ranging from various stages of chondromalacia to chondral flaps to loose bodies and fraying or tearing [15]. Femoral head decompression can relieve bone marrow edema, but decompression does not address other intraarticular changes. THA is an open surgery that involves the removal of the femoral head and pain caused by these mechanical changes, which fundamentally resolves the intraarticular injuries and replaces the friction interface. This causes marked pain relief of the hip joint on the replaced side postoperatively. The decompressed side had less pain symptoms before surgery, which was not markedly relieved postoperatively.
Thus, the patient's expectation was not met, resulting in lower postoperative satisfaction.
For the patient's hip mobility, we focused on the difference between the internal and external rotations of the decompression and replacement sides. Early symptoms of ONFH are often pain and the limitation of internal rotation [16]. Decompression surgery does not interfere with structures of the hip joint. Thus, postoperative inflammation of the hip joint still exists, and even progresses further. The preoperative internal rotation activity of the hip joint is limited and is not significantly improved by surgery. In THA, the hip joint is thoroughly cleaned and loosened, and the hip joint activity is improved. Moreover, there is more preoperative activity on the replaced side than on the decompressed side, so the 9 improvement of the hip joint activity on the replaced side is more notable. In this study, the degree of internal rotation of the hip was better on the replaced side than on the decompressed side. Therefore, we believe that this limitation in the internal rotation of the hip joint postoperatively affected the postoperative satisfaction.
We also performed a bilateral one-leg standing test to compare the time of standing on one side. In this study, compared with the replaced side, the decompressed side had a shorter standing time. When standing on one leg, the gluteal muscles (gluteus maximus, gluteus medius, and gluteus maximus) are the main supports [17]. When the strength of the gluteal muscles is insufficient on one side of the patient, their supportive ability is poor, thus reducing the standing time on one foot. It was suggested that when the hip undergoes decompression, the affected limb should not undergo weight bearing for at least 3 months. Thereafter, weight bearing should be restored based on the recovery process. However, the replacement side can resume weight bearing after surgery. The limbs on the decompression side will have different degrees of gluteal muscle atrophy, which may cause hip joint fatigue and soreness in advance due to the insufficient strength of the gluteal muscle on the decompression side. The patient's satisfaction with the surgical outcome is then reduced.
Psychological factors also have important effects on the postoperative satisfaction of patients [18]. Patients may have residual discomfort or partial dysfunction after decompression. In this study, patients still felt hip discomfort on the decompressed side when comparing their two hips at the same time. As a result, the postoperative patient satisfaction was lower for the decompressed side than for the replacement side. With the continuous improvement of living standards, patients have increasing requirements for quality of life. Although the postoperative function of the decompression side can fully meet the functional needs of the hip joint in patients' daily lives, the residual mild pain or partial dysfunction after surgery occasionally makes the patient feel uncomfortable, making some patients prefer THA over decompression.
This study has some limitations. First, the sample size of this study was relatively small, and further investigation is needed using larger samples. Second, in the study, only postoperative pain and some functions in the patients were compared. The factors we studied were too simple. Whether other factors also affect patient satisfaction with decompression surgery requires further study.

Conclusions
One-stage THA and multiple drilling decompression are effective surgical procedures for bilateral ONFH at different stages. However, a few patients with no progression on radiographic evaluation may require two-stage THA on the decompressed sides, and this may be due to more pain, poorer internal rotation, and weaker gluteal muscle strength when compared with the replaced sides. The requirement for written informed consent was waived off owing to the retrospective nature of the study.

Consent for publication
Consent to publish was attained from the study participants.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no conflict of interest.

Funding
This study was funded by the Natural Science Funding of Jilin province (Grant number: 20170414020GH) for providing support of design

Authors contributions
Every author participated in a particular role in this study. LD had the role in research design, analysis and interpretation of data and writing of manuscript. YHG participated in a particular role in editing of manuscript. SZ and YFH had the role in acquisition of data.
JGL provided facilities. XQ took responsibility for the integrity of the work as a whole, from inception to published work and acquisition of funding.   Figure 1 Kaplan-Meier survivorship analyses when THA was used as the endpoint. The survival rate of decompressed sides within 2 years was 80% in these one-stage THA and multiple drilling decompression when THA was used as the endpoint on the decompression side.
16 Figure 2 Kaplan-Meier survivorship analyses when the progression of necrosis or severe pain was used as the endpoint. The survival rate of decompression surgery in our study would be 86.7% when the progression of necrosis or severe pain was used as the endpoint Figure 3 X-ray films of preoperative and after operation. A is the radiograph of the pelvis of the patient. It can be seen that the left femoral head has collapsed and the right femoral head is completely smooth. B is the patient's frog-leg lateral views.
The combined necrotic angle = a (119.2°)+b (99.1°). C and D are the hip magnetic nucleus of the patient before surgery. In the figure, bilateral femoral head necrosis can be observed, and the left femoral head necrosis range is large.
18 E and F were reviewed X-ray films 1 year after surgery. It can be seen that the shape of the femoral head on the decompression side (right side) is intact and smooth, and the condition has not progressed further. The bone ingrowth along the needle channel can be observed (white arrow indication). And the replacement side (left side), the hip prosthesis is in good position and the bone is ingrowth.