header advert
Results 1 - 13 of 13
Results per page:
Applied filters
Content I can access

General Orthopaedics

Include Proceedings
Dates
Year From

Year To
Orthopaedic Proceedings
Vol. 105-B, Issue SUPP_3 | Pages 13 - 13
23 Feb 2023
Tay M Monk A Frampton C Hooper G Young S
Full Access

Source of the study: University of Auckland, Auckland, New Zealand and University of Otago, Christchurch, New Zealand

The Oxford Knee Score (OKS) is a 12-item questionnaire used to track knee arthroplasty outcomes. Validation of such patient reported outcome measures is typically anchored to a single question based on patient ‘satisfaction’, however risk of subsequent revision surgery is also an important outcome measure. The OKS can predict subsequent revision risk within two years, however it is not known which item(s) are the strongest predictors. Our aim was to identify which questions were most relevant in the prediction of subsequent knee arthroplasty revision risk.

All primary TKAs (n=27,708) and UKAs (n=8,415) captured by the New Zealand Joint Registry between 1999 and 2019 with at least one OKS response at six months, five years or ten years post-surgery were included. Logistic regression and receiver operating characteristics (ROC) curves were used to assess prediction models at six months, five years and ten years.

Q1 ‘overall pain’ was the strongest predictor of revision within two years (TKA: 6 months, odds ratio (OR) 1.37; 5 years, OR 1.80; 10 years, OR 1.43; UKA: 6 months, OR 1.32; 5 years, OR 2.88; 10 years, OR 1.85; all p<0.05). A reduced model with just three questions (Q1, Q6 ‘limping when walking’, Q10 ‘knee giving way’) showed comparable or better diagnostic ability with the full OKS (area under the curve (AUC): TKA: 6 months, 0.77 vs. 0.76; 5 years, 0.78 vs. 0.75; 10 years, 0.76 vs. 0.73; UKA: 6 months, 0.80 vs. 0.78; 5 years: 0.81 vs. 0.77; 10 years, 0.80 vs. 0.77).

The three questions on overall knee pain, limping when walking, and knee ‘giving way’ were the strongest predictors of subsequent revision within two years. Attention to the responses for these three key questions during follow-up may allow for prompt identification of patients most at risk of revision.


Orthopaedic Proceedings
Vol. 105-B, Issue SUPP_3 | Pages 102 - 102
23 Feb 2023
Campbell T Hill L Wong H Dow D Stevenson O Tay M Munro JT Young S Monk AP
Full Access

Contemporary indications for unicompartmental knee replacement (UKR) include bone on bone radiographic changes in the medial compartment with relatively preserved lateral and patellofemoral compartments. The role of MRI in identifying candidates for UKR is commonplace. The aim of this study was to assess the relationship between radiographic and MRI pre-operative grade and outcome following UKR.

A retrospective analysis of medial UKR patients from 2017 to 2021. Inclusion criteria were medial UKR for osteoarthritis with pre-operative and post-operative Oxford Knee Scores (OKS), pre-operative radiographs and MRI.

89 patients were included. Whilst all patients had grade 4 ICRS scores on MRI, 36/89 patients had grade 3 KL radiographic scores in the medial compartment, 50/89 had grade 4 KL scores on the medial compartment. Grade 3 KL with grade 4 IRCS medial compartment patients had a mean OKS change of 17.22 (Sd 9.190) meanwhile Grade 4 KL had a mean change of 17.54 (SD 9.001), with no statistical difference in the OKS change score following UKR between these two groups (p=0.873). Medial bone oedema was present in all but one patient. Whilst lateral compartment MRI ICRS scores ranged from 1 to 4 there was no association with MRI score of the lateral compartment and subsequent change in oxford score (P value 0.458). Patellofemoral Compartment (PFC) MRI ICRS ranged from 0 to 4. There was no association between PFC ICRS score and subsequent change in oxford knee score (P value .276)

Radiographs may under report severity of some medial sided knee osteoarthritis. We conclude that in patients with grade 3 KL score that would normally not be considered for UKR, pre-operative MRI might identify grade 4 ICRS scores and this subset of patients have equivalent outcomes to patients with radiographic Grade 4 KL medial compartment osteoarthritis.


Orthopaedic Proceedings
Vol. 105-B, Issue SUPP_3 | Pages 78 - 78
23 Feb 2023
Bolam S Tay M Zaidi F Sidaginamale R Hanlon M Munro J Monk A
Full Access

The introduction of robotics for total knee arthroplasty (TKA) into the operating theatre is often associated with a learning curve and is potentially associated with additional complications. The purpose of this study was to determine the learning curve of robotic-assisted (RA) TKA within a multi-surgeon team.

This prospective cohort study included 83 consecutive conventional jig-based TKAs compared with 53 RA TKAs using the Robotic Surgical Assistant (ROSA) system (Zimmer Biomet, Warsaw, Indiana, USA) for knee osteoarthritis performed by three high-volume (> 100 TKA per year) orthopaedic surgeons. Baseline characteristics including age, BMI, sex and pre-operative Kellgren-Lawrence grade were well-matched between the conventional and RA TKA groups. Cumulative summation (CUSUM) analysis was used to assess learning curves for operative times for each surgeon. Peri-operative and delayed complications were reviewed.

The CUSUM analysis for operative time demonstrated an inflexion point after 5, 6 and 15 cases for each of the three surgeons, or 8.7 cases on average. There were no significant differences (p = 0.53) in operative times between the RA TKA learning (before inflexion point) and proficiency (after inflexion point) phases. Similarly, the operative times of the RA TKA group did not differ significantly (p = 0.92) from the conventional TKA group. There was no discernible learning curve for the accuracy of component planning using the RA TKA system. The average length of post-operative follow-up was 21.3 ± 9.0 months. There was no significant difference (p > 0.99) in post-operative complication rates between the groups.

The introduction of the RA TKA system was associated with a learning curve for operative time of 8.7 cases. Operative times between the RA TKA and conventional TKA group were similar. The short learning curve implies this RA TKA system can be adopted relatively quickly into a surgical team with minimal risks to patients.


Orthopaedic Proceedings
Vol. 105-B, Issue SUPP_3 | Pages 17 - 17
23 Feb 2023
Tay M Stone B Nugent M Frampton C Hooper G Young S
Full Access

Source of the study: University of Auckland, Auckland, New Zealand and University of Otago, Christchurch, New Zealand

Outcomes following knee arthroplasty are typically defined as implant survivorship at defined timepoints, or revision incidence over time. These estimates are difficult to conceptualise, and lack context for younger patients with more remaining years of life. We therefore aimed to determine a ‘lifetime’ risk of revision as a more useful metric for total (TKA) and unicompartmental knee arthroplasty (UKA).

The New Zealand Joint Registry was used to identify 96,497 primary TKAs and 13,481 primary UKAs performed between 1999 and 2019. Patient mortality and revision incidence were also extracted. Estimates of lifetime risk were calculated using an actuarial lifetable method. The estimates were stratified by age and gender. Reasons for revision were categorised using previously published standardised definitions.

The lifetime risk of UKA revision was two-fold higher than TKA across all age groups (range 3.7-40.4% UKA, 1.6-22.4% TKA). Revision risk was higher for males with TKA (range 3.4%-25.2% males, 1.1%-20% females), but higher for females with UKA (range 4.3%-43.4% vs. 2.9%-37.4% for males). Revision due to infections were higher for TKA (1.5% males, 0.7% females) compared with UKA (0.4% males, 0.1% females). The increased risk in younger UKA patients was associated with higher incidence of aseptic loosening (UKA 2%, TKA 1%) and ‘unexplained pain’ (UKA 2%, TKA 0.2%).

The risk for UKA was two-fold higher than TKA, and this was partially explained by a higher proportion of revisions due to ‘unexplained pain’. For TKA, males had higher risk of revision, in contrast to UKA where females had higher risk; this gender difference was associated with higher incidence of infections with TKA. Younger age, gender and higher ASA status were also associated with increased lifetime risk of UKA revision. Lifetime risk of revision can provide a meaningful measure of arthroplasty outcomes to aid patient counselling.


Orthopaedic Proceedings
Vol. 105-B, Issue SUPP_3 | Pages 15 - 15
23 Feb 2023
Tay M Carter M Bolam S Zeng N Young S
Full Access

Source of the study: University of Auckland, Auckland, New Zealand

Unicompartmental knee arthroplasty (UKA) has benefits for patients with appropriate indications. However, UKA has a higher risk of revision, particularly for low-usage surgeons. The introduction of robotic-arm assisted systems may allow for improved outcomes but is also associated with a learning curve. We aimed to characterise the learning curve of a robotic-arm assisted system (MAKO) for UKA in terms of operative time, limb alignment, component sizing, and patient outcomes.

Operative times, pre- and post-surgical limb alignments, and component sizing were prospectively recorded for consecutive cases of primary medial UKA between 2017 and 2021 (n=152, 5 surgeons). Patient outcomes were captured with the Oxford Knee Score (OKS), EuroQol-5D (EQ-5D), Forgotten Joint Score (FJS-12) and re-operation events up to two years post-UKA. A Cumulative Summation (CUSUM) method was used to estimate learning curves and to distinguish between learning and proficiency phases.

Introduction of the system had a learning curve of 11 cases. There was increased operative time of 13 minutes between learning and proficiency phases (learning 98 mins vs. proficiency 85 mins; p<0.001), associated with navigation registration and bone preparation/cutting. A learning curve was also found with polyethylene insert sizing (p=0.03). No difference in patient outcomes between the two phases were detected for patient-reported outcome measures, implant survival (both phases 98%; NS) or re-operation (learning 100% vs. proficiency: 96%; NS). Implant survival and re-operation rates did not differ between low and high usage surgeons (cut-off of 12 UKAs per year).

Introduction of the robotic-arm assisted system for UKA led to increased operative times for navigation registration and bone preparation, but no differences were detected in terms of component placement or patient outcomes regardless of usage. The short learning curve regardless of UKA usage indicated that robotic-arm assisted UKA may be particularly useful for low-usage surgeons.


Orthopaedic Proceedings
Vol. 105-B, Issue SUPP_3 | Pages 14 - 14
23 Feb 2023
Tay M Monk A Frampton C Hooper G Young S
Full Access

Source of the study: University of Auckland, Auckland, New Zealand and University of Otago, Christchurch, New Zealand

Patient reported outcome measures (PROMs) are predictors of knee arthroplasty revision. Unicompartmental knee arthroplasty (UKA) is effective for patients with the correct indications, however has higher revision rates than total knee arthroplasty (TKA). Different revision thresholds for the procedures have been postulated. Our aims were to investigate: 1) if PROMs could predict knee arthroplasty revision within two years of the score at six months, five years and ten years follow-up, and 2) if revision ‘thresholds’ differed between TKA and UKA.

All TKAs and UKAs captured by the New Zealand Joint Registry between 1999 and 2019 with at least one OKS response at six months (TKA n=27,708, UKA n=8,415), five years (TKA n=11,519, UKA n=3,365) or ten years (TKA n=6,311, UKA n=1,744) were included. were propensity-score matched 2:1 with UKAs for comparison of revision thresholds.

Logistic regression indicated that for every one-unit decrease in OKS, the odds of TKA and UKA revision decreased by 10% and 11% at six months, 10% and 12% at five years and 9% and 5% at ten years. Fewer TKA patients with ‘poor’ outcomes (≤25) subsequently underwent revision compared with UKA at six months (5.1% vs. 19.6%, p<0.001), five years (4.3% vs. 12.5%, p<0.001) and ten years (6.4%vs. 15.0%, p=0.02). Compared with TKA, UKA patients were 2.5 times more likely to undergo revision for ‘unknown’ reasons, bearing dislocations and disease progression.

The OKS is a strong predictor of subsequent knee arthroplasty revision within two years of the score from early to late term. A lower revision threshold was found with UKA when compared with a matched TKA cohort. Higher revision rates of UKA are associated with both lower clinical thresholds for revision and additional modes of UKA failure.


Orthopaedic Proceedings
Vol. 105-B, Issue SUPP_3 | Pages 16 - 16
23 Feb 2023
Tay M Bolam S Coleman B Munro J Monk A Hooper G Young S
Full Access

Source of the study: University of Auckland, Auckland, New Zealand

Unicompartmental knee arthroplasty (UKA) is effective for patients with isolated compartment osteoarthritis, however the procedure has higher revision rates. Long-term survivorship and accurate characterisation of revision reasons are limited by a lack of long-term data and standardised revision definitions. We aimed to identify survivorship, risk factors and revision reasons in a large UKA cohort with up to 20 years follow-up.

Patient, implant and revision details were recorded through clinical and radiological review for 2,137 consecutive patients undergoing primary medial UKA across Auckland, Canterbury, Counties Manukau and Waitematā DHB between 2000 and 2017. Revision reasons were determined from review of clinical, laboratory, and radiological records for each patient using a standardised protocol. To ensure complete follow-up data was cross-referenced with the New Zealand Joint Registry to identify patients undergoing subsequent revision outside the hospitals. Implant survival, revision risk and revision reasons were analysed using Cox proportional-hazards and competing risk analyses.

Implant survivorship at 15 years was comparable for cemented fixed-bearing (cemFB; 91%) and uncemented mobile-bearing (uncemMB; 91%), but lower for cemented mobile-bearing (cemMB; 80%) implants. There was higher incidence of aseptic loosening with cemented implants (3-4% vs. 0.4% uncemented, p<0.01), osteoarthritis (OA) progression with cemMB implants (9% vs. 3% cemFB/uncemMB; p<0.05) and bearing dislocations with uncemMB implants (3% vs. 2% cemMB, p=0.02). Compared with the oldest patients (≥75 years), there was a nearly two-fold increase in risk for those aged 55-64 (hazard ratio 1.9; confidence interval 1.1–3.3, p=0.03). No association was found with gender, BMI or ASA.

Cemented mobile-bearing implants and younger age were linked to lower implant survivorship. These were associated with disease progression and bearing dislocations. The use of cemented fixed-bearing and uncemented mobile-bearing designs have superior comparable long-term survivorship.


Orthopaedic Proceedings
Vol. 94-B, Issue SUPP_XXIII | Pages 233 - 233
1 May 2012
Hohmann E Tay M Tetsworth K Bryant A
Full Access

Given their role in reducing anterior tibial translation, the recruitment patterns and viscoelastic properties of the hamstring muscles have been implicated as neuromuscular factors contributing to the ACL gender bias. Nevertheless, it is uncertain whether patterns of aberration displayed by the female neuromuscular system significantly alters the antagonist moments generated by the hamstrings during maximal effort knee extension. The purpose of the current study was to examine the effect of gender on hamstring antagonist moments in order to explain the higher ACL injury rates in females.

Eleven females (age 30.6 ± 10.1 years, mass 62.1± 6.9 kg, height 165.9 ± 4.6) and 11 males (age 29.0 ± 8.2 years, mass 78.6± 14.4 kg, height 178.5± 6.2) were recruited as subjects. Surface electrodes were placed over the semitendinosus (ST) and biceps femoris (BF) muscles of the dominant and non-dominant limbs. Each subject performed two sets of five maximal extension and flexion repetitions at 180-1. EMG, isokinetic torque and knee displacement data were sampled at 1000Hz using an AMLAB data acquisition system.

Average hamstring antagonist torque data across the range of knee flexion for female subjects was significantly higher (%Diff=24%) than for the male control subject. Statistical analyses revealed a significant main effect of gender (F = 4.802; p = 0.036).

Given that females possess a more compliant ACL and hamstring musculature, compared with their male counterparts, an augmented hamstring antagonist may represent a compensatory neuromuscular strategy to increase knee stiffness to control tibial translation and ACL strain. The results of this project suggest that it is unlikely that gender-related differences in hamstring antagonist torque is one of the predisposing factors contributing to the higher ACL injury rates in females.


Orthopaedic Proceedings
Vol. 94-B, Issue SUPP_XXIII | Pages 234 - 234
1 May 2012
Hohmann E Tay M Tetsworth K Bryant A
Full Access

Anthropometric anatomical factors may influence mechanical and functional stability of joints. An increased posterior tibial slope places the anterior cruciate ligament at a theroretical biomechanical disadvantage. An increased posterior tibial slope can potentially alter forces during landing tasks by either increasing anterior tibial translation and/or ACL loading. The purpose of this study is to investigate the relationship between posterior tibial slope and anterior cruciate ligament injuries. It is hypothesised that subjects with an ACL injury have an increased posterior tibial slope compared to a normal population.

Posterior tibial slope in 211 patients (154 male, 57 female), aged 15–49, who underwent anterior cruciate ligament reconstruction was measured using the posterior tibial cortex as reference. A matched control group was used for comparison.

The average posterior tibial slope in the ACLR population was 6.1 degrees, whilst the control group had average values of 5.4 degrees. This finding nearly reached statistical significance (p=0.057). In the male population, average values were 5.5 degrees in the ACLR group and 5.9 in the control group. This was not significant (p=0.21). However, there was a significant difference (p=0.04) in the female group. ACLR females had higher values 6.5 degrees whereas the control group had average values of 5.2 degrees.

Increased posterior tibial slope decreases the inclination of the ACL and potentially decreases vector force during dynamic tasks. We could not confirm the results of previous studies demonstrating an increased degree of posterior tibial slope in ACL injured patients. However, we demonstrated a significant difference in tibial slope in females. Based on our results, an increased posterior tibial slope is not a risk factor in males but possibly contributes to ACL injuries in females. Increased posterior tibial slope may be one of the reasons why females have a higher incidence of ACL injuries.


Orthopaedic Proceedings
Vol. 94-B, Issue SUPP_XXIII | Pages 180 - 180
1 May 2012
Hohmann E Tay M Tetsworth K Bryant A
Full Access

Previous research has shown that tunnel placement is critical in ACL reconstruction. The ultimate position of both the femoral and tibial tunnel determines knee kinematics and overall function of the knee post surgery. As with all techniques there is a definite learning curve for the arthroscopic technique. However, the effect of the learning curve on tunnel placement has been studied sparsely. The purpose of this project therefore is to investigate the effect of the learning curve on tunnel placement.

Postoperative radiographs of the first 200 anterior cruciate reconstructions with bone-tendon-bone patella tendon of a single orthopaedic surgeon performed during the first four years of independent practice were analysed for tunnel placement. Radiographs were digitalised and imported into a CAD program.

Tunnel placement both femoral and tibial antero-posterior and sagittal was assessed using Sommer's criteria. A rating scale was developed to assess overall placement. A total of 100 points indicated perfect placement. A maximum of 30 points each were allocated for sagittal femoral and tibial placement and a maximum of 20 points each were allocated for coronal placement.

Tunnel placement scores improved from 66 for the first 25 procedures to 87 for the last 25 procedures. Sagittal femoral placement (zone 1–4 with zone 1 being the preferred zone of placement) improved from an average of 1.44 to 1.08. Sagittal tibial placement (45% from anterior border of tibia) did not change significantly and remained between 42.82 t0 44.76%. Coronal femoral placement (between 10:00–11:00 o'clock for the right knee and 1:00–2:00 for the left knee) ranged from 10.45–11.15 and 12:45-1:15 o'clock respectively. This finding may be related to the transtibial tibial technique used to place the femoral tunnel. Coronal tibial placement (45% from medial tibial border) ranged from 45-46.58%.

Correct placement of the femoral and tibial bone tunnels is important for a successful reconstruction of the anterior cruciate ligament (ACL). This study demonstrated a definitive learning curve and steady improvement of tunnel placement. Whilst there was no significant improvement in sagittal placement, overall placement improved significantly.


Orthopaedic Proceedings
Vol. 94-B, Issue SUPP_XXIII | Pages 232 - 232
1 May 2012
Hohmann E Tetsworth K Tay M Bryant A
Full Access

A higher posterior tibial slope can potentially result in kinetic and kinematic changes of the knee. These changes may influence knee functionality in ACL-deficient and ACL-reconstructed subjects. The purpose of this study is to investigate the relationship between knee functionality and posterior tibial slope in ACL-deficient and ACL-reconstructed subjects.

Subjects with isolated ACL injuries and subjects who underwent ACL- reconstruction with BPTP between 18 and 24 months post surgery were included in the study. Posterior tibial slope was measured on a lateral radiograph using the posterior tibial cortex as a reference. The Cincinnati scoring system was used to assess knee functionality.

Frty-four ACL-deficient patients with a mean age of 26.6 years, and 44 ACL-reconstructed patients with a mean age of 27.2 (25–49) years were included. The posterior tibial slope in the ACL-deficient group averaged 6.10±3.57 degrees (range 0–17 degrees) and 7.20±4.49 degrees (range 0–17) in the ACL-reconstructed group. The mean Cincinnati score in the ACL-deficient subject was 62.0±14.5 and 89.3±9.5 in the ACL-reconstructed subject.

There was a moderate but non-significant correlation (r=0.47) between knee functionality and slope in the ACL-deficient subject. By dividing posterior tibial slope into intervals, a strong significant correlation (r=0.91, p=0.01) was observed between knee functionality and slope. There was a weak but non-significant correlation (r=0.24) between knee functionality and slope in the ACL-reconstructed patient. Dividing posterior tibial slope into intervals (0-4, 5-9, >10) a strong and significant correlation (r=0.96, p=0.0001) was observed between knee functionality and slope.

The results of this study suggest that subjects with a higher posterior tibial slope have higher knee functionality. This is in contrast to previous research.


Orthopaedic Proceedings
Vol. 94-B, Issue SUPP_XXIII | Pages 235 - 235
1 May 2012
Hohmann E Tay M Tetsworth K Bryant A
Full Access

A number of validated knee outcome rating scales are used to assess knee function in the ACL-deficient and ACL-reconstructed knee. These scores use a numeric system to rate findings such as pain, swelling, subjective assessment of function and level of activity.

However, it is unknown whether there is a correlation between the outcome rating scales and whether they can be used interchangeably. The aim of this study was to investigate the correlation between the four commonly used outcome rating scales (Lysholm, IKDC, Cincinnati and Tegner).

Inclusion criteria included physically active patients between the age of 18 and 35 years with isolated ACL injuries. A power calculation for sample size was performed. Selecting an alpha level of 0.05 and power value of 0.8, 24 ACL- deficient and 24 ACL-reconstructed subjects were needed to achieve adequate statistical power. Statistical analysis included the calculation of means and standard deviations for the dependant variables. Pearson's product moment correlation coefficients were used to establish the strength of the relationships.

Forty-four ACL-deficient and 24 ACL reconstructed subjects (mean age 27.0, range 16–49), with a minimum of 12 months post surgery, completed the tests. Pre-operatively, strong significant correlations (r=0.53-0.74, p=0.0001-0.001) between IKDC and the other scoring systems (Cinncinati, Lysholm and Tegner) were observed. The Lysholm score was significantly correlated to IKDC (r=0.74, p=0.0001) and Cinncinati (r=0.60, p=0.001) scores. Non-significant moderate correlations were observed between Lysholm and Tegner (r=0.38, p=0.17) and Cinncinati and Tegner (r=0.36, p=0.18) scores. Post-operatively all scores were strongly related (r=0.61- 0.93). However, only the relationships between Lysholm and IKDC score (p=0.001) and IKDC and Cinncinati score (p=0.01) reached statistical significance.

The results of this study indicate that the commonly used rating scales produce interchangeable results in the ACL-deficient patient. In the ACL-reconstructed patient, knee scoring systems seem to measure different aspects of physical activity, physical disability and subjective patient satisfaction—all of which are not interchangeable. As such, the classification of results may vary and may explain the findings.


Orthopaedic Proceedings
Vol. 94-B, Issue SUPP_XXIII | Pages 192 - 192
1 May 2012
Hohmann E Tay M Tetsworth K Bryant A
Full Access

Epidemiologic studies project an increase of hip fractures worldwide. They are an important cause of morbidity and mortality in the elderly and represent an increasing burden on a country's health service. The aim of the study was to evaluate the mortality of hip fractures admitted to a regional hospital in Australia and calculate the relative risk ratio of morbidity variables on mortality.

This retrospective review included all patients admitted from 2003 to 2008 to a regional Queensland hospital with a hip fracture. The relative risk ratio for the probability of death was calculated for the following variables: previous mobility (independent, home with help, nursing home), type of treatment (hemiarthroplasty, ORIF, DHS/Nail, total hip arthroplasty, conservative), ASA, comorbidities (dementia, hypertension, cardiac, respiratory, renal, previous hip fractures, diabetes), pre-operative haemoglobin, BUN ratio, length of stay, operative time, anaesthetic time and type (general, spinal) and, gender.

A total of 211 patients (136 female, 75 male) with an average age of 79.1 years were admitted. Seventy-six patients died during the specified interval. The average 30 day mortality was 6.2% and the average time of survival was 318 days. The relative risk of death was above one for the following variables: female gender 1,16; nursing home 1,11; more than 1 comorbidity 1,38; more than 4 comorbidities 1,78; dementia 1,12; diabetes 1,3; hypertension 1,35, previous fractures 1,43; ASA 4 1,5; operating time more than 120 minutes 7,4; length of stay more than 20 days 2,16, BUN ratio>0.1 1,38 and BUN ration<0.04 1,78.

This retrospective project identified a number of variables influencing mortality of hip fractures. These results demonstrate that the relative risk substantially increases with length of surgical time, length of hospital stay in excess of 20 days and more than four associated comorbidities.