Journal of Orthopaedics

OBJECTIVES: Percutaneous vertebral augmentation has improved the quality of life in patients with osteoporotic vertebral fractures. Our objective is to review 19 patients with these interventions and to analyze the obtained improvement.

PATIENTS AND METHODS: Retrospective study on 10 kyphoplasties, 9 vertebroplasties, 6 men, 13 women, 78.9% in the thoracolumbar junction, Genant grade I (42.1%), posterior wall involvement (21.1%), cement leakage (31.6%). We analyzed the Oswestry test, the vertebral and regional kyphosis, Cobb angle, vertebral collapse, visual analogue scale of the pain (VAS), among others.

RESULTS: With kyphoplasty we obtained a correction of the vertebral kyphosis (p=0.024) and with vertebroplasty a correction of the collapse (p=0.014), both statistically significant. Significant differences did not appear in the Oswestry test between both techniques, although there was a global improvement. The time of intervention was greater in kyphoplasties (100 +/- 44.53 minutes), as was the time of radiation (5.70 +/- 1.33 minutes) and the price (4,026.40 +/- 1,016.14 euros). The reduction of the pain (VAS) was better for kyphoplasties (9.70 +/- 1.76 to 3.50 +/- 1.84 points).

CONCLUSIONS: In our experience and despite its greater cost and greater radiation dose kyphoplasty provides a greater correction of the kyphosis and the pain than vertebroplasty.

Percutaneous vertebral augmentation is a procedure that has allowed an improvement in the quality of life in patients with osteoporotic vertebral fractures. Solely in the United States 1.5 million osteoporotic fractures per year were described in 2004, of which 700,000 were seated at the vertebral level1. Due to the elevated frequency of this pathology and to its growing increase, techniques were developed to correct these vertebral fractures in a less invasive or percutaneous way. In the decade of the 80’s Galibert2developed the first vertebroplasty to treat a haemangioma of the C2 vertebra. In 1989, Lapras applied this technique to the treatment of vertebral fractures, with encouraging results. On the other hand from the decade of the 90’s, authors such as Reiley3, developed a technique that combined the classic percutaneous augmentation of vertebroplasty with a previous raising of the vertebral collapse by means of an expansion balloon, thus arising kyphoplasty giving a greater correction of the height and the vertebral kyphosis.
Our objective is to review the interventions in 19 patients in our centre between 2004 and 2005 for vertebral fractures and 1 haemangioma, by means of vertebroplasty and kyphoplasty, analyzing the clinical, radiological and functional recovery reached by these patients.

Material and Methods :

A retrospective descriptive study on a group of 19 patients, of whom 6 were men (31.6%) and 13 women (68.4%) with an average age of 59.05 years, and a standard deviation of 12.80 years with a minimum of 33 years and maximum of 81 years, which had undergone 10 kyphoplasties (52.6%) and 9 vertebroplasties (47.4%). We analyzed:
 Personal data: name, history number, sex, technique, life previous to the intervention.
 Characteristic clinical histories prior to the intervention: Lasegue’s sign, mobility, sensitivity, reflexes, sphincters etc., and after the intervention:-
 Fracture location, region, cause, osteoporotic, time of evolution, Genant grade, type, Computerized Axial Tomography (CAT), Nuclear Magnetic Resonance (NMR), consideration of the involvement of the posterior wall and/or the medullary canal.
 Preoperative and postoperative days, time from the intervention to walking, time in the operating room, type of anaesthesia, radiation, costs, and complications.
 Vertebral and regional Kyphosis, Cobb angle, vertebral collapse.
 Visual analogue scale of the pain (VAS) before and after.
 Need for analgesics and type used.
 OPTQoL (Osteoporosis Targeted Quality of Life Questionnaire)4.
 Oswestry test.

The statistical methodology applied differed according to 2 aspects. Descriptive analysis. In the first place, a global exploration of the data was made to identify extreme values and to characterize differences between sub-groups of cases. The numerical variables were expressed with means and standard deviations or, if the distributions were asymmetric with medians and percentiles (P25-P75), and the qualitative variables with tables of frequencies and percentages. Inferential analysis. In order to evaluate the global mean differences between quantitative variables before and after the intervention, the Student t-Test for paired samples or the Wilcoxon nonparametric signed-rank test was used. In order to compare the mean values of quantitative variables the Student t-Test for independent samples or the Mann-Whitney nonparametric test was applied. When significant differences were detected they were quantified by determining confidence intervals for mean differences to 95%. In order to study the relationship between qualitative variables, the Chi-square test with continuity correction or Fisher’s exact test (for poorly populated 2x2 tables) was applied. In order to detect changes in the before and after answers due to the experimental intervention, the marginal homogeneity test, a nonparametric test for two related ordinal variables, was used. This test is an extension of the McNemar test for multivariate binary responses. Also, the McNemar nonparametric test was applied with the same objective for two related dichotomizing variables.
The analysis of the data was made with SPSS 13.5 for Windows program.

78.9% of our patients presented an active previous life, whereas 21.1% referred to a sedentary previous life. Mobility, sensitivity and reflexes were preserved in 100% of the cases. A sphincter involvement of 5.3% was notable for fractures where an invasion of the medullary canal took place, in spite of this, the remaining 94.7% remained within normality in this respect.
The most frequent fractures were those of T12 and L1, with a 31.6% frequency in each one, followed by those of L2 or L3 at 10.5% in each. A fracture of L1 was associated to another fracture in 2 cases and that of L2 in 1 case.
The thoracolumbar junction T11-L2 was involved in 78.9% of the cases, followed by the lower lumbar (L3-L5) in 15.8% of the cases, and with the thoracic (T1-T10) in 5.3% of the cases.
The causal agent was minimal trauma in 73.7% of the cases, emphasizing osteoporosis as the suspected cause in 68.4% of the cases.
With regard to the classification according to the type of fracture, we found a predominance of type I of the Genant classification, that is, with a collapse between 20-25%, in 42.1% of the cases, nevertheless, in types II or III the percentage went up to around 36.8% or 21.1% respectively. The fractures in wedge formation predominated in 52.6% of the cases, followed by the biconcave in 31.6% and those produced by crushing in the remaining 15.8%.
NMR was needed in 18 cases (94.7%), with an involvement of the posterior wall in 21.1%, with repercussion on the rachidian canal in 10.5%. Nevertheless, tests such as CAT were also needed in 7 cases, 36.8%, where an involvement of the posterior wall was also shown in 21.1%, but with this test the involvement of the rachidian canal was only demonstrable in 5.3%. All this leads us to consider that a more sensitive diagnosis of invasion of the rachidian canal would be achieved with NMR rather than with CAT, at least in our series.
With regard to the date of the intervention, 42.1% were performed between 2 months and one year (chronic-early mode), as opposed to the sub-acute ones between 2 weeks and 2 months in 26.3%, the acute ones less than 2 weeks in 21.1% and the chronic-delayed ones more than a year in the remaining 10.5%.
Regarding the OPTQoL and other data related to the quality of life prior to the intervention, we observed that 68.4% of our patients were bed-ridden, 21.1% had a bed-armchair life and 10.5% considered their life as limited by the pain. Basic daily activities were shown to be disabled in 78.9% because of the pain and 21.1% were recognized as being able but with severe pain. Some 47.4% took non-steroidal anti-inflammatory drugs (NSAIDS), 42.1% took NSAIDS and tramadol and 5.3% used morphine patches. Muscular relaxants were taken by 63.2% of our patients. Some 89.5% referred to a lot of difficulty for their physiological necessities and 10.5% of our patients noted moderate difficulty. 94.7% indicated that they had enormous difficulty in attempting to pick up an object. Up to 89.5% of our patients considered themselves housebound by the pain. 89.5% were very scared of falling, with 5.3% quite scared. Some 47.4% of our patients perceived losses of height, including 42.1% who perceived humps. The psychological repercussion of these deformities led to 15.8% needing to hide these changes. 15.8% had undergone previous fractures. In 94.7% of the cases they recognized their quality of life as bad. 42.15% recognized having members in their family with osteoporosis. 78.9% never consulted for osteoporosis. Finally, up to 94.7% had never had any bone densitometry (BDM), nor received preventive treatment before their fracture, a finding which illustrates the need to initiate a preventive treatment of osteoporosis in postmenopausal patients.
However, after the intervention 0% were bed-ridden and 10.5% referred to a bed-armchair life, which suggests a great improvement in the quality of life. 26.3% recognized a life limited by the pain and finally, in 47.4% and 15.8% they described a normal life with/without pain. 0% were disabled by the pain after the intervention for their basic daily activities, whereas 15.8% were able, but with severe pain. Postoperatively 21.1% continued taking NSAIDS, 10.5% NSAIDS and tramadol, 5.3% morphine patches and pregabalin or gabapentin was prescribed in 10.5%. Often these pains had a neuropathic component associated with spondyloarthrosis and lumbar sciatica present at the moment of the fracture. The cases in which there were cement leakages (31.6%), and especially in the cases which were symptomatic (5.3% of the cases), also required postoperative sedatives. 31.6% took muscular relaxants after the intervention. Only 10.5% referred to great difficulty performing their physiological necessities, 31.6% referred to moderate difficulty and up to 47.4% denied having problems on the matter. 21.1% described having enormous difficulty trying to pick up object, as opposed to 15.8% who did not refer to any difficulty. After the intervention only 5.3% recognized that they were housebound due to the pain. Finally the quality of life after the intervention was bad in 5.3% of the cases, fair in 52.6% and good in 42.1% of the cases. All this leads to the consideration of how the techniques of percutaneous vertebral augmentation in our centre have been able to improve the quality of life of our patients.
The mean pre-intervention VAS changed from 9.79 points with a standard deviation of 1.273 points, to a mean post-intervention VAS of 4.68 points with a standard deviation of 2.286 points. It entailed a statistically significant difference, with p=0.0, with concrete lower and upper limits (3.90;6.30). Nevertheless, we did not find statistically significant differences between kyphoplasty and vertebroplasty with regard to the reduction of VAS. What is worthy of mention was the change of the mean pre-intervention VAS of 9.70 points with a standard deviation of 1.76, to a mean post-intervention VAS of 3.50 points with standard deviation of 1.84 points for the kyphoplasty in our series, which showed a better improvement of the VAS with this technique than with vertebroplasty which gave a change of the mean pre-intervention VAS from 9.89 points with a standard deviation of 0.33 points, to a mean post-intervention VAS of 6 points with a standard deviation of 2.06 points. This smaller reduction of the VAS after vertebroplasty with respect to kyphoplasty in our series may explain that 100% of the patients submitted to kyphoplasty would recommend this procedure to other patients, as opposed to only 66.7% that would recommend vertebroplasty after having this intervention.
At the global level, 84.2% acknowledged that they would have the operation again, as opposed to 15.8% that would not have it again. Also the quality of life referred to by the patients was better after the kyphoplasty interventions: being good in 60%, as opposed to vertebroplasties where the final quality of life was good in 22.2%, and bad in 11.1%. There is the possibility of an error factor in that the age of the patients submitted to kyphoplasty was less (mean of 52.80 years with a standard deviation of 11.16 years) than that of those submitted to vertebroplasty, (mean of 68 years with a standard deviation of 11.20 years), as a conditioner of these last results.
As far as the surgical procedure the degree of anaesthetic risk conditioned a predominance of ASA II in 52.6%. The most used anaesthetic procedure was local anaesthesia with sedation in 89.5%, although general anaesthesia was required in 10.6%.
We evaluated the vertebral kyphosis, the Cobb angle, the regional kyphosis and the percentage of vertebral collapse5 prior to, and after, the operation.

We found a statistically significant difference in the correction of the vertebral kyphosis for kyphoplasties with a p=0.024 (1.35;11.54) and there was also a statistically significant difference for the correction of the vertebral collapse for vertebroplasties with a p=0.014 (3.50;21.87). (See tables 1 and 2). (See figures 5 and 6.) Although both techniques corrected both values, in our series the statistically significant differences were centred in the results given above.

We found differences in the time of radiation, the time of intervention and the price, all these parameters being greater for kyphoplasties.

Comparative table	Kyphoplasty	Vertebroplasty
Vertebral kyphosis correction	15º	7,87º
Cobb’s angle correction	11º	5,12º
Regional kyphosis correction	15,2º	7,62º
Preoperatory days	3,4 days	1,56 days
Postoperatory days	1,20 days	1,33 days
Time to walk (days)	1,20 days	1 days
Radiation time	5,70 minutes	4,55 minutes
Surgical time	100 minutes	80.56 minutes
Costs	4026,40 Euros	1436,67 Euros

Table. 1. Comparative table between both techniques. We can observe differences on costs, radiation time and surgical time.

With regard to the Oswestry test, differences were found in the results before and after the application of the surgical technique. In fact mean pre-intervention values changed from 92.27 points with a standard deviation of 6.16 points to mean post-intervention values of 43.08 points with a standard deviation of 22.19 points. Which would determine that a statistically significant difference existed, p= (0.00), (33.75-60.62), after the application of both techniques in the quality of life information gathered in the Oswestry test. Nevertheless, significant differences were not found in the results of this test when comparing both techniques, within our series.

Finally, we found some complications such as, persistent kyphotic attitude with continued pain after the treatment in 5.3%, cement leakage in 31.6% (6 cases), symptomatic in 5.3% (persistent pain) and persistence of the pain without cement leakage in 31.7%. We did not find serious complications such as the pulmonary embolisms described in the bibliography. We considered that a cement leakage had taken place when it was indicated in the clinical history by the surgeon who had made the intervention previously.

When injected at the vertebral level polymethylmethacrylate, PMMA, allows a reinforcement of the vertebral body to develop, avoiding the forces of micromovement, compression and deformation, conferring a greater strength for the support of loads and greater resistance to deformation by compression, and causes an exothermic reaction that causes damage in sensory endings, producing analgesia of the zone. Also the curing process of PMMA causes the formation of toxic monomers that damage the nociceptive endings. These thermal and/or toxic effects cause anti tumour like cytotoxic effects. In our series we evaluated its use both as a vertebral reinforcement, and as an agent which can entail analgesia of the zone by the said mechanisms. In our patients we did not incorporate antibiotics to the PMMA, in fact, the administration of gentamicin or tobramycin is known to be capable of reducing the strength of the cement up to 24%5. Nevertheless, it has been observed that it does have some clinical significance, which is why some authors continue applying these techniques with these antibiotics. Recently, the use of a new cement preparation for the treatment of osteoporotic vertebral fractures has been described, called SrHA (Strontium hydroxyapatite) it is a bioactive bone cement containing strontium, taking 15-18 minutes to cure, and reaching a polymerization temperature of up to 58º, with a capacity to support compression stresses of up to 40.9 MPa and allowing stabilization, mineralization, the induction of bone formation, as well as osteointegration6.
In our series we tried to follow criteria to indicate or to contraindicate a technique of percutaneous vertebral augmentation, in fact, and in relation to that found in the bibliography7, we indicated it in the cases of pain associated to osteoporotic vertebral fractures, benign tumour, such as aggressive and/or painful haemangioma, even in malignant, and Kümmel’s disease. Nevertheless, we did not indicate it in those cases where risks of active infection at the vertebral level existed, osteomyelitis type, diskitis, epidural abscess, intracanalicular tumoural extension, disorders of coagulation and/or compression fractures with deformity greater than 70%.
In general, it has been considered in the bibliography that vertebroplasty has been able to reduce the morbidity, the time of hospitalization and the care required after an osteoporotic vertebral fracture 8;9. Also, kyphoplasty would have the advantage over vertebroplasty in correcting the deformity of the fractured vertebral body in a controlled manner10. In our experience we have tried to compare both techniques and have seen that our results correspond with these affirmations.
In our series we found that vertebroplasty as much as kyphoplasty provided an improvement in the mean pre-intervention VAS of 9.79 points with a standard deviation of 1.27 points, to a mean post-intervention VAS of 4.68 points with a standard deviation of 2.28 points. Nevertheless, the improvement reached in VAS was better in kyphoplasties than in vertebroplasties when obtaining a change from a mean pre-intervention VAS of 9.70 points with a standard deviation of 1.76 points, to a mean post-intervention VAS of 3.50 points with a standard deviation of 1.84 points, which would be over the post-intervention VAS reached by vertebroplasty with a mean of 6 points. These results are reinforced in the consulted bibliography, since Rhyne11, for example, speaks of a change of a mean pre-intervention VAS of 9.16 points to a mean post-intervention VAS of 2.91 points after the performance of kyphoplasties on 52 patients, whereas McKiernan12 obtains a mean pre-intervention VAS of 7.7 points, with a mean post-intervention VAS of 2.8 points after the performance of 49 vertebroplasties; thus giving improvements of 6.25 points for kyphoplasties and 4.9 points for vertebroplasties. In our series we did not find statistically significant differences between both techniques regarding the VAS results, in fact other authors like Pflugmacher13, recognize that they have not found differences in VAS between 20 patients with vertebroplasty and 22 with kyphoplasty.
Neither is a large amount of PMMA needed to obtain the desired effect, thus, Guglielmi14 emphasizes a good functional recovery after the injection of between 2 and 5 ml of intravertebral cement, which would agree with our surgical technique. It must be remembered that an excess of cement or pressure, could entail the appearance of later leakages.
In our series we reached a statistically significant correction of the vertebral kyphosis (p=0.024) (1.35;11.04), with the kyphoplasties, from a mean of 15º with a standard deviation of 5º, to a post-intervention mean of 8.80º with a standard deviation of 2.77º. All this agrees with that shown by Rhyne11, with a correction of 3.4º with kyphoplasties (p<0.05). Also other authors13note a greater correction of the vertebral kyphosis with kyphoplasties with a p<0.05.
In our series we found a statistically significant correction of the vertebral collapse, changing from a mean pre-intervention vertebral collapse of 25.4% to a mean post-intervention 12.76%, p=0.014, (3.50,21.87). This supposes a correction of 12.64º. In the bibliography, we found restitutions of the prior vertebral collapse of 30% and 50% at the mean vertebral level10, with kyphoplasties correcting the vertebral collapse in 60% of the acute fractures and in 26% of the chronic fractures15. Which is in favour of kyphoplasty as it would be useful also for its correction of the collapse, although better in the acute fractures that in the chronic, in which cases it is more difficult to correct this deformity. We shared this idea and reserved this technique for acute cases rather than chronic. The correction of the vertebral collapse was for us more statistically significant in vertebroplasties than in kyphoplasties. In a self-critical mode, our sample, of about 19 cases, is not very great and possibly with a greater sample a statistically significant correction in the cases with kyphoplasty would also be reached. We also emphasize the measurement problems of these radiological parameters where several observers are used, which may cause possible bias in inter-observer individual variability. In our series the review of the images was the responsibility of 3 different observers.
With regard to the collected data of the Oswestry scale, we obtained an evolution from the pre-intervention mean of 92.27 points with a standard deviation of 6.16 points to a post-intervention mean of 43.08 points with a standard deviation of 22.19 points, which supposes an improvement of 49.19 points. This difference is significant, not only statistically speaking, with a p= (0.00), (33.75-60.62), but because according to Fairbank16, the American Food and Drug Administration has chosen a minimum difference of 15 points between the preoperative and postoperative evaluations of the Oswestry questionnaire as an indication of clinical change in patients undergoing spinal fusion. Therefore, in our series the significant improvement of these patients is stated, although we do not establish important differences between both techniques with regard to the Oswestry test. Pflugmacher13 finds significant differences between both techniques with regards to the Oswestry test results.
Finally, as in all the series, we have also had complications. Thus persistent kyphotic attitude with pain continuing after treatment appeared in 5.3%, cement leakage (6 cases) in 31.6% being symptomatic in 5.3% (with persistent pain) and the persistence of pain without cement leakage in another 31.7%. For authors like Evans17, who described a series of 488 patients submitted to vertebroplasties, the rate of complications rose to around 4.9% and it was considered that vertebroplasty is a safe and effective technique. However, there are complication rates in the bibliography, especially relating to the cement leakage, that lead to the consideration of the risks of the intervention. Thus, McKiernan12 notes a risk of leakage around 7.7%, Majd10, locates it between 20 and 65% for vertebroplasties. Nevertheless, Majd10 notes a risk of cement leakage of 10.6% for kyphoplasties and Rhyne11 places it at 9.8%. All this would lead to thinking that there is a greater leakage risk with vertebroplasties than with kyphoplasties. In fact, in our series we found 2 cases of cement leakage in the patients with kyphoplasty and 4 cases in the patients with vertebroplasty. Equally, a greater percentage of contiguous fractures have been described in vertebroplasties than in kyphoplasties, thus McKiernan12 notes a 6.5% risk after vertebroplasties, whereas Gaitanis18 considers that this risk oscillates between 10-39% for kyphoplasties and 12.4-52% for vertebroplasties. Other risks described are hypotension, costal fractures, damage in the dura mater, pulmonary embolisms, respiratory distress (ARDS), cerebral cement embolisms, radicular compression after intraforaminal cement leakage, paraplegia, cauda equine syndrome, intracanal compression, toxicity, allergy or thermal damage. Many of these effects could be explained by the multiple vascular connections established in the vertebral column, highlighting especially the internal and external vertebral venous plexuses, and the basivertebral veins 1919. Venography prior to vertebroplasty has been proposed with the intention of reducing the risks of embolisms, but in our series it was not necessary. In fact there are authors like Vasconcelos20, who recognize that the performance of venography prior to vertebroplasty is not necessary, since after 205 vertebroplasties only 3 (1.5%) complications appeared, such as hypotension, hypoestesia in the puncture zone and proximal radiculopathy, concluding that in none of the cases could it be said that it had been caused by cement leakage. Vasconcelos recommends vertebroplasty with a precise fluoroscopic control as a better preventive measure.

Conclusion:

In our experience and despite its greater cost, greater radiation dose and greater operating time, kyphoplasty produces a greater, statistically significant (p=0.024), correction of kyphosis, and of the pain than vertebroplasty. With vertebroplasty we reached a statistically significant correction of vertebral collapse (p=0.014). We reached a global improvement in the Oswestry test with both techniques, p=(0.00) (33.75-60.62), although without statistically significant differences between them.

1. Riggs BL, Melton LJ 3rd. The worldwide problem of osteoporosis: insights afforded by epidemiology. Bone.1995 Nov;17(5 Suppl):505S-511S.

2. Galibert P, Deramond H, Rosat P, Le Gars D. [Preliminary note on the treatment of vertebral angioma by percutaneous acrylic vertebroplasty]. Neurochirurgie.1987;33(2):166-8.

3. Garfin SR, Yuan HA, Reiley MA. New technologies in spine: kyphoplasty and vertebroplasty for the treatment of painful osteoporotic compression fractures. Spine.2001 Jul 15;26(14):1511-5.

4. Lydick E, Zimmerman SI, Yawn B, Love B, Kleerekoper M, Ross P, et al. Development and validation of a discriminative quality of life questionnaire for osteoporosis (the OPTQoL). J Bone Miner Res.1997 Mar;12(3):456-63.

5. Amar AP, Larsen DW, Esnaashari N, Albuquerque FC, Lavine SD, Teitelbaum GP. Percutaneous transpedicular polymethylmethacrylate vertebroplasty for the treatment of spinal compression fractures. Neurosurgery.2001 Nov;49(5):1105-14; discussion 1114-5.

6. Cheung KM, Lu WW, Luk KD, Wong CT, Chan D, Shen JX, et al. Vertebroplasty by use of a strontium-containing bioactive bone cement. Spine.2005 Sep 1;30(17 Suppl):S84-91.

7. Dixon RG, Mathis JM. Vertebroplasty and kyphoplasty: rapid pain relief for vertebral compression fractures. Curr Osteoporos Rep.2004 Dec;2(4):111-5.

8. Diamond TH, Champion B, Clark WA. Management of acute osteoporotic vertebral fractures: a nonrandomized trial comparing percutaneous vertebroplasty with conservative therapy. Am J Med.2003 Mar;114(4):257-65.

9. Papaioannou A, Adachi JD, Parkinson W, Stephenson G, Bedard M. Lengthy hospitalization associated with vertebral fractures despite control for comorbid conditions. Osteoporos Int.2001;12(10):870-4.

10. Majd ME, Farley S, Holt RT. Preliminary outcomes and efficacy of the first 360 consecutive kyphoplasties for the treatment of painful osteoporotic vertebral compression fractures. Spine J.2005 May-Jun;5(3):244-55.

11. Rhyne A 3rd, Banit D, Laxer E, Odum S, Nussman D. Kyphoplasty: report of eighty-two thoracolumbar osteoporotic vertebral fractures. J Orthop Trauma.2004 May-Jun;18(5):294-9.

12. McKiernan F, Faciszewski T, Jensen R. Quality of life following vertebroplasty. J Bone Joint Surg Am.2004 Dec;86-A(12):2600-6.

13. Pflugmacher R, Kandziora F, Schroder R, Schleicher P, Scholz M, Schnake K, et al. [Vertebroplasty and kyphoplasty in osteoporotic fractures of vertebral bodies -- a prospective 1-year follow-up analysis]. Rofo.2005 Dec;177(12):1670-6.

14. Guglielmi G, Andreula C, Muto M, Gilula LA. Percutaneous vertebroplasty: indications, contraindications, technique, and complications. Acta Radiol.2005 May;46(3):256-68.

15. Crandall D, Slaughter D, Hankins PJ, Moore C, Jerman J. Acute versus chronic vertebral compression fractures treated with kyphoplasty: early results. Spine J.2004 Jul-Aug;4(4):418-24.

16. Fairbank JC, Pynsent PB. The Oswestry Disability Index. Spine.2000 Nov 15;25(22):2940-52; discussion 2952.

17. Evans AJ, Jensen ME, Kip KE, DeNardo AJ, Lawler GJ, Negin GA, et al. Vertebral compression fractures: pain reduction and improvement in functional mobility after percutaneous polymethylmethacrylate vertebroplasty retrospective report of 245 cases. Radiology.2003 Feb;226(2):366-72.

18. Gaitanis IN, Hadjipavlou AG, Katonis PG, Tzermiadianos MN, Pasku DS, Patwardhan AG. Balloon kyphoplasty for the treatment of pathological vertebral compressive fractures. Eur Spine J.2005 Apr;14(3):250-60.Epub 2004 Oct 8.

19. Groen RJ, du Toit DF, Phillips FM, Hoogland PV, Kuizenga K Coppes MH, et al. Anatomical and pathological considerations in percutaneous vertebroplasty and kyphoplasty: a reappraisal of the vertebral venous system. Spine.2004 Jul 1;29(13):1465-71.

20. Vasconcelos C, Gailloud P, Beauchamp NJ, Heck DV, Murphy KJ. Is percutaneous vertebroplasty without pretreatment venography safe? Evaluation of 205 consecutives procedures. AJNR Am J Neuroradiol.2002 Jun-Jul;23(6):913-7.