Abstract

In the pediatric population, scoliosis is classified into congenital, syndromic, idiopathic, and neuromuscular in aetiology. Syndromic scoliosis represents a wide range of systemic anomalies associated with scoliosis. The primary challenge for a clinician is to think beyond the scoliotic curve, as the underlying pathology is multisystemic. The aim of this review is to identify the systemic anomalies, associated with syndromic scoliosis.

MEDLINE, EMBASE, and CINAHL databases were searched, dating from 1990–2020, relevant to the purpose of our study. Keywords used: “scoliosis”+ “syndrome” + “genetic”. Retrospective, prospective studies were included. Case reports that had fewer than 4 patients were not included.

Delineating 60 articles, we found a total of 41 syndromes to be associated with scoliosis. Thoracic region was the most common level of scoliosis curve, being noted in 28 syndromes. Mental retardation, seizures, and ataxia were the commonly noticed CNS anomalies. VSD, ASD, and TGA were the anomalies associated with CVS; Hypotonia, rib and vertebral malformations were the most identified neuromuscular anomalies; pulmonary hypoplasia, renal agenesis, and strabismus were other associations.

A multidisciplinary approach, involving spinal surgeons, paediatricians, geneticists, anesthesiologists, and allied health professionals, is vital for the best care of patients with syndromic scoliosis.

The location of the scoliotic curve reflects the associated anomalies, as thoracic curvature is more closely linked with cardiac anomalies, while lumbosacral curvature is seen to be often linked with genitourinary anomalies. We hope that this article provides a clear overview of the systemic associations in syndromic scoliosis and thus, facilitates and streamlines the management protocol.

Abstract

Non-operative management of AIS can present practical and psychological challenges, as effective bracing requires a considerable investment of time in adolescence which is a formative point of physical and emotional development. The management team lacks input from the psychological team and thus, it would be prudent for the spinal teams to appreciate and deal with the psychosocial effects associated with bracing.

We sought to investigate how bracing as a part of non-operative management of idiopathic scoliosis, is perceived among adolescents.

We performed a search of CINAHL, Medline, AMED, PsychARTICLES, Psychology and Behavioral Sciences Collection and PsychINFO databases to identify qualitative research investigating the thoughts, feelings and experiences and attitudes of those undergoing bracing for AIS. Keywords used were (((“adoles∗” OR “young pe∗”) AND “idiopathic scoliosis”) AND “brac∗”) AND (“perce∗” OR “experience∗” OR “perspective∗” OR “attitude∗”).

Ten research articles were identified using our search strategy. Only one article addressed our research question specifically. This reported that almost all adolescent patients experienced psychological difficulties during treatment and received most of their support from family and friends rather than health professionals.

Our interpretation of the literature on this subject has yielded three recommendations for service providers. First, the policy ought to change to ensure that psycholological support is engrained within the treatment process; second, more information and advice must be given to patients and their families; and third, health professionals should appreciate and directly address in their consultations the psychological difficulties involved in brace wearing and the need for adequate support.

Aims

Cauda equina syndrome (CES) can be associated with chronic severe lower back pain and long-term autonomic dysfunction. This study assesses the recently defined core outcome set for CES in a cohort of patients using validated questionnaires.

Summary Statement

Incorrect pedicle screw placement can lead to neurological complications. Practice outside the operating room on realistic bone models, with force feedback, could improve safety. Pedicle forces in cadaveric specimens are reported, to support development of a training tool for residents.

Inserting screws into the vertebral pedicles is a challenging step in spinal fusion and scoliosis surgeries. Errors in placement can lead to neurological complications. The more experienced the surgeon, the better the accuracy of the screw placement. A physical training system would provide residents with the feel of performing pedicle cannulation before operating on a patient. The proposed system consists of realistic bone models mimicking the geometry and material properties of typical patients, coupled with a force feedback probe. The purpose of the present study was to determine the forces encountered during pedicle probing to aid in the development of this training system.

We performed two separate investigations: [1] 15 participants (9 expert surgeons, 3 fellows and 3 residents) were asked to press a standard pedicle awl three times onto a mechanical scale, blinded to the force, demonstrating what force they would apply during safe pedicle cannulation and during unsafe cortical breach; [2] three experienced surgeons used a standard pedicle awl fitted with a one-degree of freedom load cell to probe selected thoracolumbar vertebrae of eight cadaveric specimens to measure the forces required during pedicle cannulation and deliberate breaching. A total of 42 pedicles were tested.

Both studies had wide variations in the results, but were in general agreement. Cannulation (safe) forces averaged approximately 90 N (20 lb) whereas breach (unsafe) forces averaged approximately 135–155 N (30–35 lb). The lowest average forces in the cadaveric study were for pedicle cannulation, averaging 86 N (range, 23–125 N), significantly lower (p<0.001) than for anterior breach (135 N; range, 80–195 N); medial breach (149 N; range, 98–186 N) and lateral breach (157 N; range, 114–228 N). There were no significant differences between the breach forces (p>0.1). Cannulation forces were on average 59% of the breach forces (range, 19–84%) or conversely, breach forces were 70% higher than cannulation forces.

To our knowledge, these axial force data are the first available for pedicle cannulation and breaching. A large range of forces was measured, as is experienced clinically. Additional testing is planned with a six-degree-of-freedom load cell to determine all of the forces and moments involved in cannulation and breaching, throughout the thoracolumbar spine. These results will inform the development of a realistic bone model as well as a breach prediction algorithm for a physical training system for spine surgery.

Introduction

Given the rising incidence of obesity in the adult population, it is more than likely that orthopaedic surgeons will be treating more obese patients with lumbar disc pathologies. The relationship between obesity and recurrent herniated nucleus pulposus (HNP) following microdiscectomy remains unclear.

Multiple reports suggest good outcome results following unicompartmental knee replacement (UKR). However, several authors report technically difficult revision surgery secondary to osseous defects. We reviewed clinical outcomes following revision total knee replacement for failed UKR and analysed the reasons for failure and the technical aspects of the revision surgery.

Between 2003 and 2009, thirty three revisions from unicompartmental knee replacement to total knee replacement were performed in thirty two patients at a single centre. Demographics, indications for the primary and revision procedures, details of the revised prosthesis including augments and any technical difficulties or complications were noted. Patient assessment included range of motion and the functional status of the affected knee in the form of the Oxford knee score questionnaire. Statistical analysis was performed with the Student t test.

All 33 revision knees were available for prospective clinical and radiological follow-up. The minimum duration of follow-up after revision surgery was 1 year (mean 3 years, range 1 – 7 years). The median interval between the original unicompartmental knee replacements to revision surgery was 19 months (range 2 – 159 months). The predominant cause of failure was aseptic loosening (50%). Other reasons included persistent pain (21%), dislocated meniscus (18%), mal-alignment (7%) and progression of symptomatic osteoarthritis in another compartment (4%). 18 of the 33 revision procedures required additional augments. During the revision surgery, 11 knees required a long tibial stem while 1 required a long femoral stem. 10 knees required medial tibial wedge augmentation; bone graft was used in 6 knees while a metal wedge augment was used in 4 to fill significant osseous defects. At the time of follow-up, range of movement averaged 103 degrees (range 70 – 120). The mean one year Oxford knee score, was 29 compared to 39 for primary total knee replacements performed during the same period in a comparable sample group of patients at our institute (p < 0.001). Three patients continued to have pain and two required re-revision; one for infection and one for loosening.

Aseptic loosening was the commonest mode of failure. Of the UKRs revised to TKRs, 90% were revised within 5 years. The majority of revisions required additional constructs. Oxford Knee Scores after revision surgery were inferior to those for primary TKR. The role of UKR needs to be more clearly defined.

Introduction

There has been renewed interest in the unicompartmental knee arthroplasty with reports of good long term outcomes. Advantages over a more extensive knee replacement include: preservation of bone stock, retention of both cruciate ligaments, preservation of other compartments and better knee kinematics. However, a number of authors have commented on the problem of osseous defects requiring technically difficult revision surgery. Furthermore, a number of recent national register studies have shown inferior survivorship when compared to total knee arthroplasty.

The purpose of this study was to review the cases of our patients who had a revision total knee arthroplasty for failed unicompartmental knee arthroplasty. To determine the reason for failure, describe the technical difficulties during revision surgery and record the clinical outcomes of the revision arthroplasties.

To investigate the clinical effectiveness and complications of caudal epidural steroid injections in the treatment of sciatica in patients with an MRI proven sacral tarlov cyst.

A Prospective case control study. All patients with corresponding radicular pain received a course of three caudal epidural steroid injections, two weeks apart and patients were reviewed at 3 months, 6 months and 1 year interval in a dedicated epidural follow up clinic. Data including demographics, MRI results, diagnosis and complications were documented. Outcome measures included the Oswestry Disability Questionnaire (ODQ), the visual analogue score (VAS) and the hospital anxiety and depression (HADS) score.

Overall patient satisfaction was recorded on a scale of 0-10.

38 patients with a sacral tarlov cyst were compared to a matched control group. In the sacral cyst group, mean VAS for axial pain reduced from 5.859 to 2.59 at three months (p<0.001). VAS for limb pain reduced from 6.23 to 2.53(<0.005). Mean ODI reduced from 45.49 at first visit to 21.98 at 3 months. Mean HADS also improved from 17 to 7. There was no statistical difference between the two groups. BMI did not affect the outcome in either group.

Based on our study, we conclude that presence of a sacral tarlov cyst is not a contraindication to caudal epidural steroid injection, as comparable significant improvement in both axial and limb pain in the short and intermediate periods was achieved without any major complications.

Purpose: Odontoid fractures are the most common cervical spine injuries in the elderly. Although octogenarians are the fastest growing age group, limited data exists on the natural history after they sustain odontoid fractures. Published mortality rates vary greatly, but are high enough to elicit comparisons to post-hip fracture mortality. It has also been suggested that halo-vest immobilization independently predicts mortality.

Method: All traumatic odontoid fractures (type II or III) seen at our institution between 1996 and 2008 were identified and only patients who were ≥ 80 years of age were selected. A retrospective chart review was performed for injury characteristics, comorbidities, hospitalization details, treatment regimen and documented complications. Patients were stratified using the Charlson comorbidities index. The primary outcome was mortality at one year and was identified using a provincial database.

Results: 72 cases were identified. Median age was 86 years (range 80 to 102). Patient treatment regimens included rigid neck collar, Halo vest orthosis, surgery or a combination thereof. 31% percent of the cohort (22 patients) was treated by Halo vest immobilization. Overall 1-year mortality rate was 15% (n=11) with only 1 Halo vest patient dying during this period. The majority of deaths (9 / 11) occurred in first 2 weeks following the injury.

Conclusion: Mortality rate in the octogenarian population sustaining an odontoid fracture is high and approaches the 1-year hip fracture mortality rate. The utilization of a Halo vest was not associated with increased mortality rate in our study. Optimal treatment regimens, and strategies to minimize morbidity, particularly in the early post-injury phase, necessitate further study.

Background: Aetiology of venous thromboembolism is multifactorial and thromboprophylaxis includes mechanical and chemical agents. There is no clear consensus on the choice of chemical agent in elective total hip arthroplasty (THA), although National Institute of Clinical Excellence (NICE) recommends low molecular weight heparin or fondaparinux to all patients.

Aim: The aim of our study was to define the efficacy and safety of various chemical agents currently used for venous throboprophylaxis – namely aspirin, warfarin and low molecular weight heparin in primary THA.

Methods: We retrospectively reviewed 905 consecutive patients with primary THA during an 18 month period. Medical notes were reviewed to record demographic data, inpatient and outpatient thromboprophylactic agents, total hospital stay, readmission, incidence of DVT, pulmonary embolism and death following surgery. Post-operative mobility, transfusion requirements and complications were noted. Suspected thromboembolic events were investigated with venous Doppler ultrasound scanning and CTPA.

Results: 417 (46%) patients received aspirin, 253 received enoxaparin, 190 patients had low dose warfarin and 45 patients had none or multiple agents for inpatient thromboprophylaxis. 615 patients had cemented and 290 patients received uncemented total hip arthroplasty. Patients predominantly received aspirin (61%) as outpatient prophylactic agent. 41 patients were investigated for a suspected thromboembolic event. 2 patients had DVT and 2 patients had PE. There were 3 deaths within 6 weeks, one each due to PE, sepsis and unknown cause. All 4 patients with thromboembolism were on enoxaparin for prophylaxis.

Conclusion: In our study aspirin was the preferred choice for thromboprophylaxis following total hip arthroplasty. We found that aspirin was most effective with no complications and enoxaparin was least effective. We advise the use of aspirin as the first choice drug for thromboprophylaxis as reiterated by some recent studies.

Introduction/Background: Uncemented total hip arthrolpasty [THA] was developed to achieve biological fixation of the prosthesis through integration of components into the prepared bone, especially in young patients with higher functional demands and longer life expectancy. The aim of this study was to update the results of a prospectiv series of primary cementless total hip arthroplasties after a minimum of 10 years follow-up

Materials and Methods: We reviewed 201 consecutive uncemented Bi-contact [Aesculap, Tuttlingen, Germany] THA in 177 patients. Harris hip score was recorded pre-operatively and at most recent follow up. Radiological measurements of the acetabulum for migration and radiolucency with Charnley-Lee zones, femur for stem subsistence, calcar resorption, prosthesis-bone radiolucency in the Gruen zones and hypertrophic calcification were carried out. Using revision as an end point, the cumulative survival of the prosthesis was noted. Survival analysis was performed using Kaplan-Meier curves and constructing life tables.

Results: 122 THR with a mean follow-up of 12.9 years were available for follow-up. 62 patients were dead and 17 patients were lost for follow-up. The mean age at surgery was 66.6 years (range 39–88).12 acetabula and 1 femur were revised in 11 patients due to any cause and 3 patients had indications for revision. The cumulative survival of the prosthesis with revision due to any cause was 95.42% at 10 years and 93.57% at 12 years. With revision due to aseptic loosening as an end-point, the cumulative survival was 97.8% at 10 years and 95.9% at 12 years. Comparison of worst-case curves using Log-rank test was statistically significant (p=0.0005).

Conclusion: Our series shows excellent stem results (without HA coating) with no revisions due to aseptic loosening. All stems were stable. Cup survivals were comparable to the other uncemented THA and cemented THA.

Introduction: Stiffness following total knee arthroplasty is a disabling problem resulting in pain and reduced function. Prevalence is not well defined and although various treatment modalities including manipulation, arthrolysis and revision surgery has been proposed with varying degrees of success for reduced flexion, these Methods: are deemed to be of limited value in fixed flexion deformity (FFD). There is limited literature on the natural history of FFD which is important to the decision process. The aim of our study was to evaluate the natural course of FFD following primary total knee arthroplasty.

Methods: Prospective review of a consecutive series of 1768 patients who underwent primary total knee arthroplasty over a 7 year (2001 to 2008) period. Demographic data included post-operative range of motion; type of prosthesis used, treatment modalities for stiffness and the final range of motion were recorded. FFD was defined as class 1(hyperextension to 0), Class 2 (1–10 degrees), Class 3(11–20 degrees) and Class 4(> 20 degrees).

All patients were reviewed by an independent reviewer (senior physiotherapist). All patients were followed from 6 weeks post surgery until FFD completely resolved or improved to patient satisfaction. Patients with infection, stiffness treated with manipulation or revision surgery were excluded from the study. Patients lost to follow-up were noted.

Results: Of the 1768 patients evaluated, 180 (10.2%) presented with a FFD. A total number of 18 patients were excluded from the study and 16 were lost to follow up. None (0%) were class 1, 134 (91.8%) were class 2, 10 (6.9%) were class 3 and 2 (1.4%) were class 4. The FFD group had a mean age of 60.5. Follow up period ranged from 1.3 to 63.3 months and the FFD improved from a mean of 8.16 degrees to 0.15 degrees (p< 0.001). In 94.5% patients the FFD completely resolved (i.e. < 5 deg) at a mean of 9.76 months. In the remaining 5.5% of patients, FFD improved from a mean of 16.4 to 6.9 degrees at a mean follow up of 15.5 months and was found to cause no functional deficit.

Conclusion: The overall prevalence of fixed flexion deformity is 10.2 % with only 0.7% in Class 3 and Class 4, which is comparable with the literature. The majority of patients will see a resolution of their fixed flexion deformity in less than 10 months with routine post operative physiotherapy. The small number of patients left with a residual FFD did not appear to suffer a functional deficit. Patients found to have a post operative FFD should be reassured and encouraged to participate in a standardised post operative physiotherapy regime.

Ruptures of the patellar and/or quadriceps tendon are rare injuries requiring immediate repair to re-establish knee extensor continuity and to allow early motion. Ultrasound is extensively used as a diagnostic tool before surgery on acute traumatic tears of the patellar tendon and quadriceps tendons.

Our aim was to re-evaluate the value/role of sonography in diagnosing quadriceps and patellar tendon rupture and in differentiating partial from complete tears. To correlate the intra operative findings with the ultrasound report.

A retrospective review of 32 consecutive patients who had a surgical intervention for suspected acute quadriceps and patellar tendon rupture over the last 3 years. Intra-operative findings, clinical, x-ray, ultrasound and MRI reports were correlated.

Seventeen patients had a suspected patellar tendon rupture on clinical examination and 15 patients had suspected quadriceps tendon rupture. Diagnosis was confirmed by clinical examination and x-rays alone in 9 patients, with additional ultrasound in 18 patients and with MRI scan in 5 patients. There were 6 false positives out of 18 [33.3%] in the ultrasound proven group and 1 false positive out of 9 [11.1%] in the clinical examination and x-ray only group. MRI was 100% accurate [n=5].

This is a small, but important study. Ultrasound offers a low degree of sensitivity and specificity in diagnosing acute quadriceps and patellar tendon ruptures. As a result, patients are being exposed to the risks of surgery unnecessarily. We recommend MRI scan as a first line of investigation in patients where the diagnosis is clinically ambiguous. Ultrasound assessment should not be relied upon in when making the decision to operate.

Pyogenic haematogenous spinal infection in the elderly, described as spondylodiscitis, vertebral osteomyelitis and epidural abscess is considered a rare but life threatening condition. Our objective was to test the hypothesis that low index of suspicion leads to delayed diagnosis and referral for definitive treatment resulting in increased and perhaps avoidable medical morbidity, social drift including early mortality and to analyse pathological entities, complications and optimum treatment options.

We performed a retrospective review of medical records over 10-year period. Post-operative infections and patients under 65 years old excluded. Initial presentation, investigations and differential diagnosis, time to diagnosis, date and day of referral, mode of definitive treatment, pathologic entities, complications and outcomes were noted. Patient outcomes were measured as duration of treatment, length of hospital stay, complications, ambulatory status, complications, discharge destination and death. Outcomes were correlated with delayed diagnosis and referral.

46 elderly (age> 65) patients with a mean age of 71 years (range=65–91). 62% referral from physicians. Fever with malaise associated with chronic LBP was the commonest presenting complaint. There were 31 patients with discitis, 12 epidural abscess and 3 osteomyelitis. Lumbar spine was affected in 63% patients. Time to diagnosis ranged from 2–17 days with mean of 8 days. Mean referral time was 9 days with 39% referrals on Friday. Staphylococcus aureus (47%) was the commonest organism isolated. Duration of hospitalisation ranged from two to twelve weeks. 46% required surgical decompression with four cases of related mortality during acute hospital stay.

Time duration to spinal referral had direct correlation with increase in morbidity, social drift and mortality. The incidence of haematogenous spinal infection in the elderly has increased over the years in our series, contrary to popular belief. A high index of suspicion in elderly patients with PUO promotes early diagnosis and optimises outcome.

Very few reports exist regarding use of intra-operative autologenous transfusion in adult spinal fusion surgery. Specific indications for use of cell saver in thoraco-lumbar spinal instrumentation are not clearly determined. Our objective was to identify the clinical factors associated with increased risk of intra-operative blood loss. To analyse the safety and benefits of using cell saver and to determine the cost-effectiveness.

51 consecutive thoraco-lumbar instrumented fusion over 14 months reviewed. There was no randomisation. Cell saver group consisted of 25 patients and control group consisted of 26 cases. Patients with degenerative scoliosis and tumours excluded. Clinical notes were reviewed.

Demographic data were comparable between the two groups. Iliac crest bone graft harvested in 20% of cell saver group and 40% in control group. Levels of fusion ranged from 1–7 (Mean=2.8) in the cell saver group and 1–4 (Mean=1.9) in the control group. Mean duration of surgery was 15 minutes longer in study group. Intra-operative blood loss was higher in cell saver group (mean=1245 mls vs 800 mls). Significant indicators for duration of surgery were number of levels fused (p< 0.0001), patient weight (p=0.003) and revision surgery (p=0.029). Significant indicators for predicting blood loss were number of levels fused (p< 0.001) and duration of surgery (p=0.0304). 20% in study group (8 units of red cells) and 26% in control group (17 units of red cells) required blood transfusion. Percentage drop in the post-operative haematocrit was 19.1 in study group compared to 36.3 in control group.

In conclusion, 44.38% blood salvaged (35–38% in spinal literature). Use of cell saver significantly decreased post-operative need for blood transfusion. Number of levels of fusion, duration greater than 4 hrs and a low pre-op Haemoglobin/Haematocrit were significant parameters in predicting intra-operative blood loss. If blood loss is less than 700 mls, gains from cell saver are debatable.

Background: Better postoperative pain, functional outcomes and patient satisfaction have been reported using spinal anaesthesia when compared with general anaesthesia.

However, higher levels of urinary retention with spinal anaesthesia can lead to delayed discharge in microdiscectomy surgery.

Anecdotally, we believe that further improvements to patient satisfaction and a reduction in the need for urinary catheterisation can be found in patients receiving fentanyl intrathecally as opposed to morphine.

Methods: Seventy consecutive patients were recruited and prospectively randomised into two equal groups, with half the patients receiving intrathecal fentanyl with the spinal anaesthetic and the remainder receiving morphine.

A comprehensive post operative evaluation was carried out documenting any anaesthetic complications, post operative analgesic requirement, physiological and functional recovery, need for urinary catheterisation and patient satisfaction.

Results: Both groups were equally matched for age and gender. Mean age was 43 years in fentanyl group and 50 years in the morhhine group. All patients were discharged on day one post surgery. 3 patients in the fentanyl group and 11 patients in the morphine group required urinary catheterisation.

No intra-operative anaesthetic or surgical complications were noted.

Mean Visual Analogue score for pain was lower in the fentanyl group [2.46] compared to morphine group [2.70].

Conclusion: Lumbar spinal microdiscectomy can be safely performed as a short stay procedure under spinal anaesthesia using intrathecal fentanyl or morphine. Post-operative pain and functional out comes were comparable between the two groups but lower incidence of post-operative nausea, vomiting, itching and urinary catheterisation can be expected with use of fentanyl

Background: The management of radicular pain due to lumbar or sacral nerve root compromise remains controversial.

Caudal epidural steroid injections are widely employed although there is little hard evidence to confirm their efficacy. This empirical treatment still remains a matter of personal choice and experience.

Objectives: To investigate the clinical effectiveness of caudal epidural steroid injections (CESIs) in the treatment of sciatica and to identify potential predictors (clinical subgroups) of response to CESIs.

Main outcome measures: The primary outcome measure was the Oswestry Disability Questionnaire (ODQ). The Visual analogue score (VAS) and the Hospital Anxiety and Depression Scores (HADS) were also employed in all cases to measure pain relief, physical and psychological function.

Method: Prospective study. All patients with corresponding radicular pain received a course of three caudal epidural steroid injections, two weeks apart.

A standard mixture of 80 mgs of triamcinalone plus 7 mls of 1% lignocaine plus 5 mls of 0.9% saline used for all patients.

Results: In the largest single series to date, we report on 628 consecutive patients, with 3 months follow up.

58 % were females, 24% smoked and 4.1% had ongoing litigation due to their pain.

Conclusion: Significant improvement in both axial and limb pain in the short and intermediate terms was achieved facilitating onward referral for physical therapy, which is fundamental in optimising outcomes.

Long term follow-up is underway.

Background: Pyogenic haematogenous spinal infection in the elderly, described as spondylodiscitis, vertebral osteomyelitis and epidural abscess is still considered a rare but life threatening condition.

Objective: To test our hypothesis that low index of suspicion leads to delayed diagnosis.

Late referral for definitive treatment may result in increased and perhaps avoidable medical morbidity, social and psychological drift, including early mortality.

Method: Retrospective review of medical records over 10-year period.

Patient pool obtained from theatre records, radiology and coding departments.

Results: Single largest series [n=46] of elderly [age> 65] patients with pyogenic spinal infections to our knowledge.

Age ranged from 65–91 with mean of 71.

Conclusions: Time duration to Spinal referral had direct correlation with increase in morbidity, social and psychological drift, and mortality.

The incidence of haematogenous spinal infection in the elderly has increased over the years in our series, contrary to popular belief.

A high index of suspicion in elderly patients with PUO promotes early diagnosis and optimises outcome.