Efforts to improve the survival of total hip arthroplasty implants have focused on alternative bearing surfaces in order to decrease wear and osteolysis1-4. The use of ceramics as bearing surfaces has had a long, successful history3,5-9. Refinements in the manufacturing process and improvements in component design have greatly reduced the material-specific complications such as component loosening and ceramic fracture10-13. The modern generation of ceramic-on-ceramic total hip arthroplasty implants became available for widespread use in the United States in March 2003. Early reports have demonstrated excellent clinical and radiographic results without the catastrophic failures associated with earlier designs4,12,14,15. However, a new finding associated with the ceramic-on-ceramic bearing as well as with other "hard-on-hard" bearings is an audible squeak16-21. The incidence of squeaking associated with ceramic-on-ceramic bearing surfaces has been reported to be <3% in North America16,20,21. Our hypothesis was that the incidence of squeaking is an underreported phenomenon.
The purpose of the present study was to better define the incidence and characteristics of squeaking in patients with ceramic-on-ceramic bearings and to compare these findings with those in a matched control group of patients who had been managed with metal-on-polyethylene bearings. In addition, clinical and radiographic outcomes were evaluated in order to gain additional insight into the possible causes of squeaking.
From March 2003 through May 2005, 783 total hip arthroplasties were performed at our institution by three surgeons (C.S.R., A.S.R., J.A.R.). Of these, 159 procedures (143 patients) were ceramic-on-ceramic total hip arthroplasties that were performed with use of the alumina V40 femoral head and Trident alumina insert (Stryker Orthopaedics, Mahwah, New Jersey)22. One hundred and forty-nine hips (131 patients) had at least one year of follow-up. The indications for the use of a ceramic-on-ceramic bearing were based on the patient's age, activity level, and preference after informed consent. All patients underwent total hip arthroplasty through a posterior approach23. No components were cemented. The femoral component that was implanted in all cases was the Accolade TMZF, and the acetabular shell was the Trident PSL Cup.
Institutional review board approval was obtained for the present study. The patients with ceramic-on-ceramic bearings were identified through a search of our total hip replacement database. The average age of the patients at the time of surgery was fifty-two years (range, twenty-three to seventy-eight years). Seventy hip replacements were performed on the left side, and eighty-nine were performed on the right side. Degenerative joint disease was the predominant diagnosis, but there were three cases of osteonecrosis, five cases of hip dysplasia, and six cases of inflammatory arthropathy (including two cases of ankylosing spondylitis, three cases of juvenile rheumatoid arthritis, and one case of rheumatoid arthritis). There was one case of nonunion of a femoral neck fracture. The average body mass index was 27.16 (range, 18.24 to 44). The study group included forty-eight women and eighty-three men.
A control group was selected in order to compare the incidence and types of noises made by total hip arthroplasty implants and the outcomes of these procedures. The control group was found by matching the patients in the ceramic-on-ceramic cohort with those in a database of patients who had received a chromium-cobalt-on-highly-crosslinked-polyethylene bearing (Crossfire; Stryker Orthopaedics). Matching criteria included age (within five years), sex, body mass index (±3), and the presence of a unilateral or bilateral hip replacement. The limiting factor in matching was the age difference between the general population in the total hip database and the younger patients with ceramic-on-ceramic bearings. As a result, forty-eight patients with sixty metal-on-polyethylene total hip replacements were identified. The patients in the matched group included twenty-two women and twenty-six men with an average age of fifty-three years (range, twenty-eight to seventy years) and an average body mass index of 27 (range, 18 to 47.2).
Preoperative and postoperative pain and function were graded according to the Hospital for Special Surgery (HSS) hip score, which rates pain, walking ability, range of motion of the hip, and function. A maximum score of 40 (10 points in each category) can be achieved24. The HSS score was determined six weeks postoperatively and at the time of the latest follow-up. Preoperatively, a patient-assessment questionnaire was used to evaluate the level of pain and function, the activity level, and patient satisfaction (see Appendix). The degree of difficulty that the patient had with activities of daily living was also assessed. The questionnaire was again administered at regularly scheduled follow-up visits at three months, six months, one year, and annually thereafter. If the patient was unable to come in for a visit, the interview was completed by telephone. One additional question was added to the postoperative questionnaire: "Does your hip make a noise?" If the answer was "yes," the patient was asked to describe the noise (for example "squeak," "grind," or "pop"), the frequency of the noise, and the activities that were associated with the noise. The patient was also asked if pain was associated with the noise and how much the noise affected his or her quality of life.
Additional investigations were performed for the patients who reported squeaking. In each case, the patient was asked when the squeaking had begun and whether the squeaking was progressive. If the patient stated that the squeaking was reproducible, he or she was videotaped attempting to reproduce the squeak. This was done to identify any pattern in movements that produced the squeak and to identify the range of motion during which the squeaking occurred.
Radiographic analysis was performed by digitizing anteroposterior radiographs with a high-resolution optical scanner (VIDAR Systems, Herndon, Virginia). The images were then analyzed with use of the semi-automated computer-assisted dual-circle technique (Hip Analysis Suite) as described by Martell and Berdia25. Both preoperative and postoperative radiographs were analyzed. The inclination of both the acetabular component and the native acetabulum of the patient were measured. The limb-length discrepancy was also measured in both groups with use of the known diameter of the cup as a reference for magnification.
Statistical Analysis
The independent-samples Student t test was used to compare the two groups in terms of the cup abduction angle, the cup anteversion angle, and limb-length discrepancy. The chi-square test was used to study the distribution of the different head diameters (28, 32, and 36 mm) in the two populations. In addition, a matched-pair analysis between the two groups was performed with use of variability limits of ±2° for anteversion and abduction angles and of ±2 mm for limb-length discrepancy. The level of significance was set at p < 0.05. The Fisher exact test was used to assess for any differences between the groups with respect to the incidence of any noises as well as that of squeaks. The Pearson correlation coefficient was used to determine if a correlation existed between self-reported daily squeaking and a negative effect on quality of life.
Source of Funding
No external sources of funding were utilized for this study.
One hundred and thirty-one patients (149 hips) with ceramic-on-ceramic bearings were available for follow-up. The average duration of follow-up was 1.83 years (range, 1.00 to 3.31 years). Forty-three (32.8%) of these patients reported that the hip made some sort of noise (Tables I and II). Specifically, fourteen patients reported an audible squeak, fourteen reported "clicks," ten reported "pops," two reported a "grind," two reported "snaps," and one patient listed "other" but provided no additional description. The most commonly reported activity that elicited noise was walking (seven patients). The remaining activities were fairly evenly distributed between exercise, bending, stair-climbing, and "other." The "other" category included write-in descriptions, including "putting on pants" for two patients and "during sex" for one patient. One patient reported a pop in the hip when moving from sitting to standing.
The fourteen patients who reported an audible squeak in association with the ceramic-on-ceramic total hip implant were further evaluated with regard to the time to the onset of symptoms, the reproducibility of the squeak, and the effect of the squeak on the quality of life. All patients reported that the squeak was gradual in onset, beginning between six and eighteen months postoperatively. No patient reported pain during the squeak, and no consistent pattern of activities was noted in association with the squeak. However, all four of the patients who were able to reproduce the squeak in the clinical setting did so by stepping up on a low stool. As the weight-bearing hip was extended from 60° of flexion to 0° or full extension, the hip squeaked (see Appendix). The combination of weight-bearing and movement through this physiologic arc of motion was required to reproduce the squeak (Fig. 1).
One of the patients who reported a squeak also had recurrent dislocations and underwent revision because of instability. An exchange of the head and liner was performed at another institution, with conversion to a metal-on-polyethylene bearing. The acetabular shell and the femoral component were retained. At the time of the latest follow-up, the patient had not had any additional noises or dislocations. Another patient was considering revision to a metal-on-polyethylene bearing because the quality of life had been adversely affected by almost daily squeaking.
The eleven patients with squeaking of the hip who had the longest follow-up were asked to further stratify their symptoms. Six patients reported that the hip squeaked at least once per day, two stated that it squeaked several times per week, two indicated that it squeaked once per week, and one reported that it squeaked "rarely." In addition, the patients rated the effect that the squeaking had on their lives on a scale of 0 to 10 (with 10 indicating a substantial negative effect on quality of life and 0 indicating no effect). The average score was 3.3 (range, 0 to 8). Five patients reported a score of =5; all of those patients reported that the hip squeaked at least once a day. Of the six patients who reported a score of between 0 and 4, only one reported that the hip squeaked daily. There was a positive correlation (r = 0.64, p = 0.035) between a hip that squeaked at least once a day and the negative effect that hip squeaking had on the quality of life.
A cohort of the patients who had a ceramic-on-ceramic bearing was directly compared with a cohort of patients who had a metal-on-polyethylene bearing (Table III). Both cohorts included forty-eight patients (sixty hips). A noise related to the total hip replacement was reported by ten patients (21%) in the ceramic-on-ceramic cohort, compared with two patients (4%) in the metal-on-polyethylene cohort; this difference was significant (p = 0.014). The two patients in the matched metal-on-polyethylene group reported the noises as "clicks." No patients in the metal-on-polyethylene group reported the noises as "squeaks," whereas three patients in the ceramic-on-ceramic group reported the noises as "squeaks."
The mean modified HSS hip score for the 131 patients with a ceramic-on-ceramic total hip implant improved from 19.8 preoperatively to 38.4 (of a possible 40) at the time of the latest follow-up. For the fourteen patients with an audible squeak, the mean HSS score improved from 20.6 to 38.2. Ninety-five percent of all patients, including those who had squeaking of the hip, reported a satisfaction score of =8 (with 10 representing complete satisfaction and 0 representing complete dissatisfaction). The only patient with a hip squeak who gave the procedure an unsatisfactory rating was the patient with chronic dislocations who underwent revision.
Radiographic Analysis
The group with squeaking (n = 14) had an average lateral opening of the acetabular component (and standard deviation) of 43° ± 5° (range, 36° to 52°) and an average anteversion of 20° ± 7° (range, 11° to 32°). The group without squeaking (n = 117) had an average lateral opening of the acetabular component of 42° ± 6° (range, 22° to 55°) and an average anteversion of 16° ± 7° (range, 3° to 38°). The average limb-length discrepancy was 2.2 ± 3.3 mm (range, -2.4 to 6.6 mm) in the group with squeaking and 1.9 ± 4.6 mm (range, -24.4 to 15 mm) in the group without squeaking. With the numbers available, there was no significant radiographic difference between the two groups. There was an identical distribution of cup inclination and anteversion between the groups with and without squeaking. Each of the fourteen patients in the group with squeaking was successfully matched with at least one patient in the control group with identical cup abduction, anteversion, and limb-length discrepancy who did not have squeaking.
In the group with squeaking, seven hip replacements had a 36-mm head and the other seven had a 32-mm head. In the overall group of ceramic-on-ceramic hip replacements, forty-eight hips (32%) had a 36-mm head, ninety-seven hips (65%) had a 32-mm head, and four hips (2.7%) had a 28-mm head, for a total of 149 implanted hips. No significant difference in the distribution of 36-mm heads was demonstrated between the hips with squeaking and all hips with a ceramic-on-ceramic implant.
Complications
Three (1.9%) of the 159 hips dislocated, one anteriorly and two posteriorly. Two were treated with closed reduction and brief immobilization and remained asymptomatic. One patient (described above) had recurrent dislocations and underwent revision. There were no cases of sciatic nerve palsy. There was one case of persistent thigh pain. One patient reported groin pain at the end of his golf swing. There were no cases of deep infection. Two patients complained of limb-length discrepancy, but only one required a shoe insert to compensate for a difference of 1 cm.
Squeaking at the site of a total hip arthroplasty is an uncommon phenomenon that is peculiar to hard-on-hard bearing surfaces. It has been described as occurring in association with both metal-on-metal and ceramic-on-ceramic bearings16-20. In addition, the literature also includes a case report of squeaking in association with a polyethylene insert26. However, that case was associated with catastrophic wear of the polyethylene insert and the squeaking occurred as the result of articulation of the head with the metal backing of the acetabular cup26. This was not the case in the current study as there was no radiographic evidence of failure. According to the manufacturer's recommendations, the optimal cup positioning for the particular ceramic-on-ceramic total hip arthroplasty implant that we used is 45° of inclination and 20° of anteversion. Achieving appropriate anteversion with ceramic-on-ceramic bearings is essential as there is no option for a ceramic elevated-lip liner. With the numbers available, we could not identify a correlation between squeaking and cup inclination or version.
The design of the ceramic liner that was used in the present study incorporates a metal backing with an elevated rim22. It is designed to protect the ceramic component from impingement and to decrease the risk of chipping and fracture during insertion, and both of these complications have been reduced with the introduction of the elevated rim22. However, it is generally recognized that a certain percentage of asymptomatic total hip replacements will impinge during normal activities. While impingement has been postulated to be the cause of the squeak18, we do not believe that the squeak is the actual sound of impingement occurring because impingement occurs at the extremes of motion and the squeaking tended to occur during a physiologic range of motion.
A mismatch of the bearing surfaces has been proposed as the etiology of a squeak that occurs between an alumina insert and a zirconia femoral head19. A second theory, proposed by Rieker et al., is that inefficient lubrication between the two bearing surfaces leads to the squeak effect in certain cases20. Those authors postulated that this deficiency of lubrication could lead to resonance or a so-called slip stick mechanism in squeaking metal-on-metal total hip implants20. The effect of lubrication on the ceramic-ceramic interface is a possible causative factor. Metal debris generated as a result of impingement may be another etiology for squeaking. The debris that is generated could serve as a third body between the two hard ceramic surfaces, possibly affecting wear and fluid film lubrication.
An association also has been made between so-called "stripe wear" and squeaking. Taylor et al. demonstrated, with hip simulators, that squeaking can be reproduced when motion and load occur along the long axis of the stripe27. Impingement, microseparation, and edge-loading also have been hypothesized as possible sources of stripe wear. Additional study, including the retrieval analysis of squeaking hips, is warranted. In the present series, no hip squeaked earlier than six months postoperatively, indicating a possible so-called run-in period before the bearing surfaces were affected.
In the present study, we found a significant difference between the ceramic-on-ceramic group and the metal-on-polyethylene group with respect to the overall incidence of noises. Noises other than squeaking may be related to the so-called hard-landing of the hard-on-hard bearing. In the case of a total hip replacement that has any distraction during the gait cycle, the impact of the two ceramic surfaces at heel strike could be interpreted by the patient as a click, pop, or snap. Other noises might also be related to soft-tissue impingement or shifting, such as with a snapping iliotibial band.
The reported incidence of squeaking following ceramic-on-ceramic total hip arthroplasty in North America has ranged from 0% to 3%16-19,21. We found an incidence of squeaking of >10% on the basis of a patient-administered questionnaire. One patient (<1%) reported squeaking before the administration of the questionnaire, and four patients (3%) were able to reproduce the squeak clinically. This finding is in accordance with the incidence of squeaking of 2.7% as reported recently by Restrepo et al.21. We were able to elicit a significantly higher incidence of squeaking with use of a patient-assessment questionnaire, which has been shown to be a reliable and reproducible method of measuring outcomes28-30.
Patients who specifically complained of hip squeaking were those in whom the hip squeaked on multiple occasions throughout the day. Many of the patients who did not complain of squeaking prior to being given the questionnaire reported that the hip squeaked no more than once a day; the squeaking did not affect either the short-term outcome or the quality of life. On the basis of our experience with the use of the questionnaire, it is likely that the incidence of squeaking has been underreported in other studies. Recently, one report from The Netherlands documented a 21% incidence of reproducible squeaking in association with ceramic-on-ceramic bearings with use of direct questioning, further suggesting that this phenomenon is underreported31.
We continue to use ceramic-on-ceramic bearings for young, active patients; however, the patients are counseled accordingly with regard to the risk of fracture and squeaking. The natural history of the squeaking hip is unknown. As some studies have suggested that early failure is a concern, we have given all of our patients with squeaking total hip replacements the option of revision to a metal-on-polyethylene or a ceramic-on-polyethylene articulation; however, to date, none have opted for revision specifically because of squeaking. Currently, we have no evidence to suggest that squeaking will lead to implant failure, but this may change as we follow these patients prospectively.
Note: The authors thank Joanne Weiskopf, MS, PA-C, for all of her assistance with the preparation of this study.