|Anterior decompression and fixation versus posterior reposition and semirigid fixation in the treatment of unstable burst thoracolumbar fracture: prospective clinical trial
|Stancic M F, Gregorovic E, Nozica E, Penezic L
This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn.
Anterior decompression and fixation (ADF) were compared with posterior reposition and semirigid fixation (PRSRF) using hook/rod with pedicle screw fixation, in the treatment of unstable burst thoracolumbar fracture in emergency circumstances. Patients with neurological injuries were operated on within the first 8 hours, whilst those without neurological deficit were operated on in the first 2 days after the fracture.
Economic study type
The study population comprised patients with incomplete neurological injury, unstable thoracolumbar burst fractures and mechanical instability associated with burst fracture and without a neurological deficit. Radiographic evidence of instability was a vertebral height loss of greater than 50% on lateral radiography, kyphosis over 20 degrees and/or spinal canal encroachment greater than 40% on axial computed tomography. Patients were excluded if they were classified as grade A according to the American Spinal Injury Association/International Medical Society of Paraplegia (ASIA/IMSOP), with no motor or sensory function preserved in the S4 and S5 sacral segment. They were also excluded if late admittance to the hospital prevented surgical intervention within an 8-hour time window.
The setting was secondary care. The economic study was carried out in Pula, Croatia.
Dates to which data relate
The effectiveness data were collected from 1 September 1997 to 31 August 2000. The cost data appear to have been collected during the same period. The price year was not stated.
Source of effectiveness data
The effectiveness data were derived from a single study.
Link between effectiveness and cost data
The cost data appear to have been collected prospectively using the same sample population as that used in the effectiveness analysis.
The sample size was not determined in the planning phase of the study to assure a certain power. Patients with unstable burst thoracolumbar fracture that were admitted to the hospital during the study period were considered at analysis. In total, 28 patients were recruited. Three patients were excluded because they did not meet the inclusion criteria. Twenty-five patients received either ADF (n=13) or PRSRF (n=12). The authors did not report any evidence that the study sample was representative of the study population.
This appears to have been a non-randomised controlled trial, which was performed at a single centre. A neurosurgeon allocated the patients to one of the groups, according to the availability of implants in the operating theatre. The duration of follow-up was generally 12 months. Six patients (3 from each group) had a shorter follow-up (longer than 6 months) since they were operated on later than the other patients. There was no loss to follow-up. The neurologist that assessed the outcomes at 6 and 12 months appears to have been blinded to the patient operative group. Different doctors, who worked independently, assessed the other effectiveness outcomes. This approach may have helped in blinding the outcome assessment. The neurosurgeon who was not involved in the outcome assessment analysed the data.
Analysis of effectiveness
All of the patients included in the study were accounted for in the analysis. The primary health outcomes assessed for both the ADF and PRSRF groups were:
the mean (and range of) neurological improvement, that is, the difference between ASIA/IMSOP grade on admission and at 12 months' follow-up;
the economic activity status in terms of the number of patients working or active, with or without limitations at 12 months' follow-up; and
the functional status, in terms of the number of patients with occasional or episodic pain, or with no pain at all at 12 months' follow-up.
The economic activity status and the functional status were assessed using the Prolo economic and functional rating scale.
The secondary end points assessed in the clinical study included blood loss, the cosmetic result for postoperative scars (scale ranging from 1 for poor to 5 for excellent results), the number of complications and the number of patients with donor site pain.
There were no statistically significant differences between the patient groups in terms of their age, gender, fracture level, fracture type (Denis classification of burst fractures) and the number of patients with ASI/IMSOP grade E (normal motor and sensory function) at admission.
The mean neurological improvement was 1.5 (range: 1 - 2) for ADF patients versus 1.57 (range: 1 - 3) for PRSRF patients, (p=0.856).
Eight of the 10 patients in the ADF group were working or active, with or without limitations, versus 8 out of 9 in the PRSRF group, (p=0.542).
Nine of the 10 patients in the ADF group experienced occasional or episodic pain, or no pain at all, versus all 9 patients in the PRSRF group, (p=0.526).
The blood loss was statistically significantly greater for ADF patients than for PRSRF patients. The average blood loss was 1,362 mL (range: 1,150 - 1,500) in the ADF group versus 750 mL (range: 500 - 1,100) in the PRSRF group, (p<0.001).
The cosmetic results for postoperative scars were better for PRSRF patients (average score = 4.1; range: 3 - 5) than for ADF patients (average score = 3.4; range: 2 - 5), (p=0.024).
There were 2 complications in the ADF group versus 1 in the PRSRF group, (p=0.531). Two ADF patients and no PRSRF patients experienced donor site pain, (p=0.260).
Neurological improvement, economic states and functional outcomes did not differ significantly between the ADF and PRSRF patients. However, PRSRF presented better results in terms of shorter operative time and hospital stay, smaller blood loss, and better cosmetic results for postoperative scars.
Measure of benefits used in the economic analysis
No summary measure of health benefit was used in the economic analysis. The study was therefore categorised as a cost-consequences analysis.
The resource quantities and the costs were not reported separately. The direct costs considered at analysis were those of the hospital, although the authors did not specify the cost categories included (only data about the duration of hospital stays were reported). The source of the direct costs was not reported. Consequently, it could not be stated whether the prices were estimated from actual data. Discounting did not appear to have been performed, although it was not relevant since the period considered at analysis was shorter than 2 years. The study reported the average costs. The price year was not stated.
Statistical analysis of costs
The mean costs and their ranges were reported. The costs were compared using the Mann-Whitney U-test, which was appropriate for the study question.
No indirect costs were reported.
No sensitivity analyses were performed.
Estimated benefits used in the economic analysis
See the 'Effectiveness Results' section.
The average hospital costs were $2,700 per ADF patient (range: 2,400 - 2,850) versus $2,250 per PRSRF patient (range: 1,900 - 2,450), (p<0.001).
It could not be inferred whether the costs associated with the complications were or were not included in the costing.
Synthesis of costs and benefits
Not applicable due to the cost-consequences approach undertaken.
Both anterior decompression and fixation (ADF) and posterior reposition and semirigid fixation (PRSRF) were equally effective in terms of neurological improvements, economic status and functional outcomes. However, PRSRF showed better results in terms of shorter operative times, smaller blood loss, absence of donor site pain, shorter hospital stays and lower hospital costs.
CRD COMMENTARY - Selection of comparators
None of the alternative health technologies under evaluation was stated to be the comparator. ADF appears to have been the most used health technology in the authors' setting. However, the authors commented that, although PRSRF was the preferred strategy, the implants were unavailable due to a lack of funds. You should decide whether these health technologies are widely used in your own setting.
Validity of estimate of measure of effectiveness
The analysis used a non-randomised controlled trial. This type of study has great potential for selection bias. The authors showed that there were no statistically significant differences between ADF and PRSRF patients in terms of their age, gender, fracture level, fracture type, and the number of patients with normal motor and sensory function at the beginning of the study. Thus, confounding may be unlikely. However, the lack of statistically significant differences between the groups may have been due to the size of the sample used in the effectiveness analysis. The small sample size may have had insufficient power to detect statistically significant differences between both groups of patients. The authors did not show any evidence that the study sample was representative of the study population. Some effort was made to assess the effectiveness outcomes blindly.
Validity of estimate of measure of benefit
The authors did not derive a measure of health benefit. The analysis was therefore categorised as a cost-consequences study.
Validity of estimate of costs
The reporting of the costs was very brief and the authors did not state the cost categories they included in the economic analysis. Therefore, it could not be easily inferred whether all the costs relevant to the perspective adopted were included. This makes it difficult to interpret the results. The costs and the quantities were not reported separately and the price year was not stated. These issues hinder reflation exercises to other settings. A statistical analysis was performed to compare the total hospital costs per patient for the ADF and PRSRF groups. Discounting was not performed, but it was irrelevant since the costs were incurred during one year.
The authors did not make appropriate comparisons of their findings with those from other studies. The issue of the generalisability of the results was not addressed. The authors' conclusions reflected the scope of the analysis.
Implications of the study
The authors recommended that a randomised controlled trial, with a proper supply of the necessary implants, should be conducted within 5 years.
Stancic M F, Gregorovic E, Nozica E, Penezic L. Anterior decompression and fixation versus posterior reposition and semirigid fixation in the treatment of unstable burst thoracolumbar fracture: prospective clinical trial. Croatian Medical Journal 2001; 42(1): 49-53
Other publications of related interest
Danisa OA, Shaffrey CI, Jane JA, Whitelhill R, Wang GJ, Szabo TA, et al. Surgical approaches for the correction of unstable thoracolumbar burst fractures: a retrospective analysis of treatment outcome. Journal of Neurosurgery 1995;83:977-83.
Gertzbein SD, Crowe PJ, Fazl M, Schwartz M, Rowed D. Canal clearance in burst fractures using the AO internal fixator. Spine 1992;17:558-60.
Kaneda K, Abumi K, Fujiya M. Burst fractures with neurologic deficit of the thoracolumbar-lumbar spine. Results of anterior decompression and stabilization with anterior instrumentation. Spine 1984;9:788-95.
Kostuik JP. Anterior fixation for burst fractures of the thoracic and lumbar spine with or without neurological involvement. Spine 1988;13:286-93.
Schenee CL, Ansell LV. Selection criteria and outcome of operative approaches for thoracolumbar burst fractures with and without neurological deficit. Journal of Neurosurgery 1997;86:48-55.
Subject indexing assigned by NLM
Adolescent; Adult; Bone Plates; Bone Screws; Chi-Square Distribution; Emergency Treatment /methods; Female; Follow-Up Studies; Fracture Fixation, Internal /instrumentation /methods; Fracture Healing /physiology; Humans; Laminectomy /methods; Lumbar Vertebrae /injuries /surgery; Male; Middle Aged; Prospective Studies; Sensitivity and Specificity; Spinal Fractures /diagnosis /surgery; Thoracic Vertebrae /injuries /surgery; Treatment Outcome
Date bibliographic record published
Date abstract record published