Stable fixation of the IBP humeral component implanted without cement in total elbow replacement: a radiostereometric analysis study of 16 elbows at two-year follow-up.

We determined the short-term clinical outcome and migration within the bone of the humeral cementless component of the Instrumented Bone Preserving (IBP) total elbow replacement in a series of 16 patients. There were four men and 12 women with a mean age at operation of 63 years (40 to 81). Migration was calculated using radiostereometric analysis. There were no intra-operative complications and no revisions. At two-year follow-up, all patients showed a significant reduction in pain and functional improvement of the elbow (both p < 0.001). Although ten components (63%) showed movement or micromovement during the first six weeks, 14 (88%) were stable at one year post-operatively. Translation was primarily found in the proximal direction (median 0.3 mm (interquartile range (IQR) -0.09 to 0.8); the major rotational movement was an anterior tilt (median 0.7° (IQR 0.4° to 1.6°)). One malaligned component continued to migrate during the second year, and one component could not be followed beyond three months because migration had caused the markers to break off the prosthesis. This study shows promising early results for the cementless humeral component of the IBP total elbow replacement. All patients had a good clinical outcome, and most components stabilised within six months of the operation.

Do tibiofemoral contact point and posterior condylar offset influence outcome and range of motion in a mobile-bearing total knee arthroplasty?

PURPOSE: The posterior condylar offset (PCO) and the tibiofemoral contact point (CP) have been reported as important factors that can influence range of motion and clinical outcome after total knee arthroplasty. A mobile-bearing knee implant with an anterior posterior gliding insert would in theory be more sensitive for changes in PCO and CP. For this reason, we analysed the PCO and CP and the relation with outcome and range of motion in 132 patients from a prospectively documented cohort in this type of implant. METHODS: The prosthesis used was a posterior cruciate retaining AP gliding mobile-bearing total knee replacement (SAL II Sulzer Medica, Switzerland). In 132 knees, the pre- and postoperative PCO and postoperative CP were evaluated. Measurements were made on X-rays of the knee taken in approximately 90° of flexion and with less than 3-mm rotation of the femur condyles. The outcome parameters, range of motion (ROM) and the knee society score (KSS), for each knee were determined preoperatively and at 5-year follow-up. RESULTS: The mean KSS improved from 91 to 161 at 5-year follow-up (p < 0.001) and the mean ROM from 102 to 108 (p < 0.05). The mean PCO difference (postoperative PCO-preoperative PCO) was--0.05 mm (SD 2.15). The CP was on average 53.9% (SD 5.5%). ROM was different between the 3 PCO groups (p = 0.05): patients with 3 or more mm decrease in PCO had the best postoperative ROM (p = 0.047). There was no statistical difference between the postoperative ROM between patients with a stable PCO and those with an increased PCO. There was no correlation between the difference in PCO and the difference in ROM; R Pearson = -0.056. There was no difference in postoperative ROM or postoperative total KSS between CP <60% and CP >60%: p = 0.22, p = 0.99, for ROM and KSS, respectively. Scatter plots showed uniform clouds of values: increase or decrease in PCO and CP had no significant influence on ROM or KSS. CONCLUSION: The hypotheses that a stable PCO and a more natural CP increase postoperative ROM and improve clinical outcome could not be confirmed. On the contrary, a decreased PCO seemed to improve knee flexion. Furthermore, a relationship between PCO and CP could not be found. LEVEL OF EVIDENCE: Prospective cohort study, Level II.

The self-aligning knee prosthesis: clinical and radiological outcome and survival analysis of a cruciate retaining meniscal bearing knee at 10-year follow-up.

PURPOSE: Previously, the clinical and radiological results of the self-aligning cruciate retaining total knee prosthesis at 5-years follow-up were described. The 10-year follow-up is now reported to determine the clinical and radiological results and the survival of the prosthesis. METHODS: A 10-year follow-up was performed in this prospectively followed cohort of 246 (232 patients) self-aligning, cruciate retaining total knee arthroplasties performed between 1995 and 1998. Because patients had died and others suffered from comorbidities making them unable to attend a follow-up consultation, data could be obtained for 105 prostheses. Clinical results were evaluated by the Knee Society Score, and radiological follow-up was performed on standard AP and lateral X-rays. All radiographs were evaluated for the presence of radiolucencies, osteolysis and wear. RESULTS: At the 10-year follow-up, there were no new cases of revision so that the cumulative survival rate at 10 years remains 95%. The 10-year follow-up shows good clinical results with an average Knee Society Score of 154 (SD 35.7) and an average range of motion of 109° (SD 13.7). After excluding 25 prostheses for comorbidities, the mean 10-year KSS score increased from 154 to 163 (SD 26.2). On the other hand, for the ROM, the exclusion of the 25 prostheses had no influence on the mean ROM value of 109°. At the 10-year follow-up, none of the 100 available radiographs showed any radiolucencies >2 mm on any component. CONCLUSIONS: Long-term follow-up of the mobile bearing, cruciate retaining, SAL-II total knee arthroplasty shows excellent clinical and radiological results without bearing dislocation. LEVEL OF EVIDENCE: Prospective cohort, Level IV.

Ultra sonographic findings for chronic lateral epicondylitis.

PURPOSE: To assess which individual gray-scale and color Doppler US findings and their combination are strongly associated with lateral epicondylitis. Also to determine whether chronic lateral epicondylitis is possible without any positive US findings. METHODS: 49 patients (6 bilateral) underwent gray-scale ultrasonic imaging between 2005 to 2007. All had a history of lateral epicondylitis and had concordant pain during US probe compression in the common extensor region. Mean patient age was 47 (sd 7.7) years; M/F ratio 21/28; L/R ratio 17/32. Five symptom free volunteers (all bilateral) with a mean age of 36 (sd 8.7) years; M/F = 4/6; L/R = 5/5. RESULTS: Neovascularity determined by color Doppler and four gray-scale US findings - a convex external contour, an erosive lateral epicondular cortex, internal calcifications, or a tear - have a specificity and PPV of 100% with conclusive likelihood ratios. However, only the sensitivity for neovascularity is above 50%. A combination of gray-scale and color Doppler shows a sensitivity between 92% to 100%, a 90% specificity with a 98% PPV and a high likelihood ratio (9 to 10). CONCLUSION: The combination of gray-scale and color Doppler changes is diagnostically superior to identify chronic lateral epicondylitis. Signs which confirm the diagnosis are a convex boundary, an erosive cortex, internal calcifications, a tear, and neovascularity. Patients with positive clinical signs and concordant pain but no US findings require further MRI evaluation.

Foot Build Registration System (FBRS) to evaluate foot posture: a reliability study with healthy subjects and patients with Charcot-Marie-Tooth disease.

BACKGROUND: To follow the weight bearing foot posture in subjects, a measurement tool using digital photography was developed, Foot Build Registration System (FBRS) (Fig. 1) [M. Van der Cruijsen, Voetvorm registratie systeem, Boxmeer (1999).]. This study's objective was to investigate the reproducibility of FBRS measurements in healthy feet and feet of patients with Charcot-Marie-Tooth disease (CMT). METHODS: Reproducibility and reliability studies were performed in several foot views in healthy and CMT patients. RESULTS: These studies showed that the variability of the 95% prediction limit depended upon the foot view being studied and whether markers had been drawn. Some individuals had a higher intra-individual variability than others. Limiting data collection to those individuals with a SD<3.5 degrees for a series of five or more photographs per view improved the 95% prediction limits. These varied between 2.8 degrees and 7.7 degrees. CONCLUSIONS: If the differences found between registration are greater than the abovementioned, values can be attributed to time or operative management for healthy and CMT patients and not to measurement error.

The Self-Aligning knee prosthesis: clinical and radiological outcome and survival analysis at 5-year follow up.

The purpose of this study is to evaluate the clinical and radiological results and the survival of the SAL II mobile bearing knee prosthesis at 5-year follow up. Between February 1995 and March 1998 a total of 246 SAL II total knee arthroplasties were performed in 232 patients. The mean follow up was 5.0 years. Thirteen revisions had been carried out (5%). The mean total Knee Society Score increased from 88 preoperatively to 153 and 155 at 1 and 5 years follow up, respectively. The mean flexion angles were 106 degrees and 107 degrees pre- and postoperatively. None of the prostheses showed radiological loosening. No dislocations or subluxations of inserts were seen. The Kaplan-Meier cumulative survival is 95% for revision for any reason at 5.0 years, with a worst case scenario of 91%. The survival for aseptic loosening is 99%. The results of SAL II after a minimum follow up of 5 years are favourable and comparable with fixed bearing and other mobile bearing designs in terms of Knee Society Score and survival. Noteworthy are the good radiological results of this device which showed a good fixation of the prosthesis at 5 years.

A modified tibial tubercle osteotomy for patellar maltracking: results at two years.

An abnormal lateral position of the tibial tuberosity causes distal malalignment of the extensor mechanism of the knee and can lead to lateral tracking of the patella causing anterior knee pain or objective patellar instability, characterised by recurrent dislocation. Computer tomography is used for a precise pre-operative assessment of the tibial tubercle-trochlear groove distance. A distance of more than 15 mm is considered to be pathological and an indication for surgery in symptomatic patients. In a prospective study we performed a subtle transfer of the tibial tuberosity according to the information gained from the pre-operative CT scan. This method was applied to two groups of patients, those with painful lateral tracking of the patella, and those with objective patellar instability. We evaluated the clinical results in 30 patients in each group. The outcome was documented at 3, 12 and 24 months using the Lysholm scale, the Kujala score, and a visual analogue pain score. Post-operatively, all but one patient in the instability group who had a patellar dislocation requiring further surgery reported good improvement with no further subluxation or dislocation. All patients in both groups had a marked improvement in pain and functional score. Two patients sustained a tibial fracture six and seven weeks after surgery. One patient suffered a per-operative fracture of the tibial tubercle which later required further fixation. If carefully performed, this type of transfer of the tibial tubercle appears to be a satisfactory technique for the treatment of patients with an increased tibial tubercle-trochlear groove distance and who present with symptoms related to lateral maltracking of the patella.

The uncemented, titanium-coated rm cup: Survival and analyses of failures.

This study evaluates the incidence of revision of the monoblock, uncemented, titanium-coated RM (Robert Mathys) cup (Mathys Medical, Bettlach, Switzerland) in primary total hip arthroplasty. Between 1991 and 1995 a total of 630 RM cups were implanted in 462 patients. A Kaplan-Meier survival analysis was performed with failure defined as acetabular revision for any reason. The mean follow-up was 8.2 years. A total of 34 acetabular revisions (5%) were carried out: 15 for recurrent dislocation due to cup malposition, nine for wear caused by impingement of a loose femoral component, five for aseptic loosening, only one cup was revised for articular polyethylene wear, and four for other causes. The cumulative survival rate of the RM cup is 91% at 10.7 years, with a worst case scenario of 87%. The cumulative survival rate for aseptic loosening was calculated: 99% at 10.7 years. The titanium-coated RM cup shows an impressive survival rate with revision for aseptic loosening as endpoint. Correct cup positioning is the key to success and requires experience and meticulous surgical technique. (Hip International 2005; 15: 71-7).

The Zweymller cementless total hip prosthesis in patients aged 50 years and younger.

Sixty-eight young patients (74 prostheses), mean age 43 years (standard deviation 7.7) at the time of the index operation, with a Zweymller cementless total hip prosthesis were evaluated with a mean follow-up time of 75 months (sd 15.7). At follow-up, the mean Harris Hip Score was 94 (sd 8.1). Six stem prostheses had vertical sub-sidence and nine cups showed signs indicating possible loosening. The Oxford Hip Score averaged 19 points (sd 8.7). Revision surgery was performed for septic loosening (n=3 stem and cup prostheses), for aseptic loosening of the cup (n=2), and for a traumatic periprosthetic fracture (n=1 stem). Worst case survival analysis for aseptic loosening of the cup showed a probability for revision of 3% (CI95 0 7.2%) with a cumulative survival of 96% (CI95 100 90.3%) after 84 months. For the stem prostheses the probability for revision was 1% (CI95 0 4.3%) after 114 months with a cumulative survival of 99% (CI95 100 95.8%) after 72 months. Worst case cumulative survival for any reason of revision was 94.5% (CI95 99.9 89.2%) after 81 months. Promising results of the Zweymller cementless hip prosthesis for the younger patient were obtained, although longer follow-up will be necessary. (Hip International 2005; 15: 1-11).

Single or double-level anterior interbody fusion techniques for cervical degenerative disc disease.

BACKGROUND: The number of surgical techniques for decompression and solid interbody fusion as a treatment for cervical spondylosis has increased rapidly, but the rationale for the choice between different techniques is unclear. OBJECTIVES: The goal of this study was to determine which method of anterior cervical interbody fusion at a single or double-level provides the best clinical and radiological outcome in patients with degenerative disc disease. SEARCH STRATEGY: Studies were identified with a computer-assisted search of electronic databases in the Cochrane Central Register of Controlled Trials (Issue 1, 2004), MEDLINE (1966 to 2004), EMBASE (1980 to 2004), and Current Contents (1996 to 2004). We also searched references of selected articles. SELECTION CRITERIA: With the aid of a checklist, two reviewers independently screened the identified references. Consensus was reached through negotiation. A third reviewer was consulted if consensus could not be reached. Inclusion criteria included: articles were reports of randomised comparative studies; treatments compared anterior cervical decompression and interbody fusion techniques, participants were individuals scheduled for surgery for a chronic (longer than 12 weeks) diagnosis of degenerative disc disease. DATA COLLECTION AND ANALYSIS: Methodological quality was assessed independently by two reviewers, using the van Tulder list of criteria. With the aid of a data extraction form, data was extracted independently by two reviewers on group characteristics, intervention details and outcome measures. MAIN RESULTS: Fourteen studies with 939 patients evaluated three comparisons of different fusion techniques. From these comparisons it appears that discectomy alone has a shorter operation time, hospital stay, and post-operative absence from work than discectomy with fusion, while there is no statistical difference for pain relief and rate of fusion. It also appears that fusion techniques that use autograft give a better chance for fusion than interbody fusion techniques that use a cage, but other outcome variables could not be combined. REVIEWERS' CONCLUSIONS: The low quality of the trials prohibits extensive conclusions from this review. More studies with better methodology and reporting are needed. There should be a more general agreement between researchers on which outcome parameters should be used in the evaluation of anterior cervical fusion procedures.

CT density measurement of bone graft within an intervertebral lumbar cage: increase of hounsfield units as an indicator for increasing bone mineral content.

Radiographic assessment of lumbar interbody fusion is difficult, especially in the presence of cages and metal artifacts. The purpose of this study is to assess whether impacted bone chips in intervertebral lumbar cages show a tendency toward increased attenuation on postoperative computed tomography (CT) scans as a sign of ongoing revascularization and increased bone mineral content inside the cage. Twenty-one patients with single- or double-level lumbar fusion underwent CT scanning 1-44 months postoperatively. CT attenuation of bone inside the cage was measured with Hounsfield units. Intra- and interobserver variability were evaluated. Regression analysis showed an increase of 7.5 HU/month postoperatively for an initial value of 615 HU. Intra- and interobserver variability showed an interclass coefficient of 0.97. CT attenuation of bone graft inside an intervertebral cage increases in the postoperative period. Hounsfield unit measurement can be performed with a high degree of accuracy and reproducibility and may in the future provide a useful tool in studying cage contents in individual patients.

Total hip arthroplasty with Boneloc cement: Unsatisfactory results in 163 hips after 9 to 11 years.

Boneloc bone cement was introduced in the Netherlands in 1992. Inferior short-term results were reported which led to the withdrawal of Boneloc from clinical use in 1995. However, little is known about the long-term outcome of hip arthroplasties with Boneloc. Between April 1992 and August 1994, Boneloc was used in 163 Mallory-Head primary total hip arthroplasties in 163 patients. Follow-up analysis was performed in 2003. To date, 27 hips (17%) have been revised for aseptic loosening of the femoral component. Median time to revision was 5.5 years. Survival analysis based on revision for aseptic loosening showed 77% cumulative survival at 11 years. With revision for aseptic loosening and/or definite radiological loosening according to Harris as endpoint, cumulative survival was 59% at 11 years. In 27 of 43 patients with definite radiological loosening, a cement fracture was seen at a median of 2.9 years. These results show failure of Boneloc cemented total hip arthroplasties occurring even during the later follow-up. Continuing periodic clinical and radiological examination is recommended. (Hip International 2004; 14: 229-32).

Patients with lateral tracking patella have better pain relief following CT-guided tuberosity transfer than patients with unstable patella.

In patients with either lateral tracking patella or unstable patella the pathological lateral position of the tuberosity can be corrected by a medial transfer. This study compared the results of subtle CT-guided correction of the tuberosity for objective unstable patella (n=27) with the results for lateral tracking patella (potential instability) as described by Dejour (n=16). Follow-up was 37 months. CT revealed a pathological lateralization of the tibial tuberosity-trochlear groove greater than 15 mm in 41 knees. These patients underwent medialization of the tibial tuberosity up to 10-12 mm lateral from the trochlear groove, and 28 patients underwent a distalization to normalize the Caton index to 1.0-1.2. Results were evaluated using Cox' method. Patients with objective patellar instability were rated as 11% excellent, 52% good, 33% fair, and 4% poor. All patients became stable except one who had a 6 degrees valgus alignment. Although 96% had improved stability, 33% of the patients still had pain. The patients with lateral tracking patella (potential instability) were rated as 37.5% excellent, 44% good, and 19% fair. The lower proportion of pain relief in patients with unstable patella is likely the result of the cartilage damage experienced by these patients following multiple dislocations. Thus the patient with lateral tracking patella without patella dislocations must be differentiated from the one with unstable patella. Their prognosis in pain relief is better.

Teardrop-femoral head distance after shelf acetabuloplasty for Perthes' disease.

Shelf acetabuloplasty is an accepted treatment for Perthes' disease. Favourable outcomes have been reported. This study aimed at determining whether changes in femoral head subluxation after shelf acetabuloplasty occur and whether they predict outcome. We assumed that reduction of subluxation after shelf acetabuloplasty would improve the long-term outcome. During follow-up, growth presented as a source of error when comparing the amount of absolute subluxation. Correction for growth was achieved by calculating a subluxation index. This index showed a significant reduction during follow-up. There was no correlation between clinical and radiological outcome, but the fact that the subluxation index showed a decreasing trend could be promising regarding outcome. This might be one of the reasons why shelf acetabuloplasty produces its effect. Following shelf acetabuloplasty the enlarged acetabulum could possibly direct the growth of the femoral head to a better-contained joint. This could result in better joint congruity and lesser tendency to degenerative arthritis.

Posterior reduction and anterior lumbar interbody fusion in symptomatic low-grade adult isthmic spondylolisthesis: short-term radiological and functional outcome.

The aim of this study was to evaluate the short-term radiological and functional outcome of surgical treatment for symptomatic, low-grade, adult isthmic spondylolisthesis. Twelve patients underwent a monosegmental fusion for symptomatic spondylolisthesis. Posterior reduction with pedicle screw instrumentation was followed by second-stage anterior interbody fusion with a cage. All patients underwent a decompressive laminectomy. At an average of 2.1 (range 1.4-3.0) years following surgery, all patients completed the Oswestry questionnaire, VAS back pain score and a questionnaire detailing their work status. Radiographs were evaluated for maintenance of reduction and fusion. The patients (nine male, three female; mean age 42, range 22-54 years) had experienced preoperative symptoms for an average of 38 (range 6-96) months. An average preoperative slip of 21% (range 11-36%) was reduced to 7% (range 0-17%). Reduction of slip was maintained at latest follow-up, at which time the average VAS score was 2.8 (range 0-8) and the average Oswestry score was 13 (range 0-32). All patients achieved a successful fusion. There were no postoperative nerve root deficits. All patients stated that they would be prepared to undergo the same procedure again if required. Seventy-five percent returned to their pre-symptom work status. Our findings suggest that posterior reduction and anterior fusion for low-grade adult isthmic spondylolisthesis may yield good functional short-term results. A high fusion rate and maintenance of reduction with a low complication rate may be expected. Further follow-up is necessary to evaluate long-term outcome.

Bioenergetics in cardiac hypertrophy: mitochondrial respiration as a pathological target of NO*.

A rat aortic banding model of cardiac hypertrophy was used to test the hypothesis that reversible inhibition of mitochondrial respiration by nitric oxide (NO*) elicits a bioenergetic defect in the hypertrophied heart. In support of this hypothesis, the respiration of myocytes isolated from hypertrophied hearts was more sensitive to exogenous NO* (IC(50) 200 +/- 10 nM vs. 290 +/- 30 nM in controls, P = 0.0064). Hypertrophied myocytes also exhibited significantly elevated inducible NO* synthase (iNOS). Consistent with this endogenous source for NO*, the respiration of hypertrophied myocytes was significantly inhibited at physiological O(2) tensions versus controls. Both the nonspecific NOS inhibitor nitro-L-arginine and the iNOS-specific inhibitor N-[3-(aminomethyl)- benzyl]acetamidine. 2HCl reversed this inhibition, with no effect on respiration of control myocytes. Consistent with an NO*-mediated mitochondrial dysfunction, the ability of intact perfused hearts to respond to a pacing workload was impaired in hypertrophy, and this effect was reversed by NOS inhibition. We conclude that endogenously generated NO* can modulate mitochondrial function in the hypertrophied heart and suggest that this bioenergetic defect may underlie certain pathological features of hypertrophy.

Elevated potassium intake inhibits neointimal proliferation in the swine coronary artery.

BACKGROUND: Previously, we reported that elevated extracellular potassium concentration in vitro inhibited proliferation and migration of vascular smooth muscle cells, formation of free radical compounds by macrophages, and reduced platelet sensitivity to agonists. More recently, we described a reduction in neointimal proliferation after balloon angioplasty injury in the carotid arteries of rats associated with an elevation of dietary potassium intake during a 4-week experiment. In the present study we conducted a similar investigation in the swine coronary artery balloon angioplasty model. PROCEDURES: Two groups of seven castrated male swine were studied; for 28 days the normal potassium group consumed a diet containing 0.25% potassium and the high potassium group ate diet containing 2.0% potassium. After 14 days on the diet, balloon angioplasty was performed. After an additional 14 days on the same diets the hearts were removed, and normal and lesioned sections of the artery were analyzed histologically. RESULTS: The neointimal area was markedly less in the high potassium group than in the normal potassium group, 0.33+/-0.04 mm2 v 0.74+/-0.10 mm2 (P < .004). Neointimal area-to-total wall area ratio in the normal potassium group averaged 0.199+/-0.018, significantly greater than the ratio computed for the elevated potassium group, 0.120+/-0.015 (P < .006). CONCLUSION: These results support the hypothesis that a high level of dietary potassium intake inhibits neointimal proliferation after balloon angioplasty in the swine coronary artery.

Is there a relation between neuropsychologic variables and quality of life after stroke?

OBJECTIVES: To describe the quality of life (QOL) of stroke patients and to distill neuropsychologic predictors for poor QOL. DESIGN: A cohort study in which patients were neuropsychologically assessed at a mean of 72.2 days after stroke, with follow-up at a mean of 9.8 months after stroke. SETTING: Research department of a rehabilitation center. PATIENTS: A consecutive sample of 164 stroke patients (mean age, 55.2yr) recruited from a university hospital, a regional hospital, and a rehabilitation center. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Orientation, memory, attention and concentration, visuospatial and visuoconstructive functions, language, and arithmetic skills were assessed with neuropsychologic tests. QOL was assessed with the Sickness Impact Profile (SIP). RESULTS: An overall mean SIP score +/- standard deviation of 20 +/- 11 showed that stroke has a high impact on everyday functioning. Further analyses indicated that QOL is related in particular to tests measuring spatiotemporal and/or sequential aspects of behavior. Forward/backward stepwise regression analysis (n = 106) showed that poor QOL was more likely if patients had a poor result on the Trailmaking Test (TMT) B and/or were women. CONCLUSION: The predictive value of the TMT is most effective and very useful because the TMT is a short and economical procedure. However, the gender-related aspects of recovery deserve more attention, as does the possible bias that can be caused by the composition of a measurement. Further research is needed to refine predictive models that are needed to facilitate the development of more adequate, individual rehabilitation programs.

Shelf acetabuloplasty for Perthes' disease: 12-year follow-up.

The goal of all therapies for Perthes' disease is to achieve an optimal shape of the acetabulum and an optimal coverage of the femoral head. Thirty patients who were included in this follow-up study (mean follow-up 12 years) underwent a shelf acetabuloplasty for Catterall group III or IV. The mean IOWA hip score at follow-up was 96 (max. 100 points, range 74-100). The mean acetabular-head quotient increased from 82.9% pre-operatively to 107.9% postoperatively, and remained 102.4% at follow-up. There was a decrease in mean lateral subluxation ratio from 1.44 pre-operatively to 1.27 postoperatively, which remained 1.23 at follow-up. According to the Stulberg classification in the 18 adult hips, 6 hips had a good result (Stulberg 1 or 2), 10 hips had a fair result (Stulberg 3), and 2 hips had a poor result (Stulberg 4 or 5). These results appear to be better than the natural history as described by Stulberg. Shelf acetabuloplasty can be considered as an appropriate surgical treatment for severe cases of Perthes' disease.