Percutaneous kyphoplasty (PKP), mainly including transpedicular approach and transverse process-pedicle, was a common minimally invasive surgical technique for the treatment of OVCF due to its outstanding therapeutic effect and simple manipulation. Transpedicular approaches of PKP were frequently discussed and compared in the preview related literatures [15-18]. A multi-center retrospective study conducted by Yilmaz et al. [19] found that significantly shorter surgery time, less bone cement amount, lower complication incidence but similar radiographical correction and symptom relief were noticed with unilateral transpedicular PKP when compared with bilateral transpedicular approach. Another study conducted by Tang et al. [20] also showed that unilateral trans-pedicular PKP had similar surgical efficiency but shorter operative time, lower hospital cost and less radiation exposure. Although the consensus has been reached by preview studies that unilateral pedicular approach PKP had many advantages, only limited amount of researchers recommended it as a preferred treatment for OVCF. It might be contributed to the following defects of unilateral transpedicular approach[21, 22] : First of all, asymmetrical vertebral restoration caused by unilateral balloon dilatation and asymmetrical distribution of bone cement would elevate the risk of contralateral vertebral collapse, resulting in the formation of wedge-shaped vertebral body and finally iatrogenic scoliosis. Next, bone cement leakage was prone to take place due to the limited volume of unilateral distribution. Furthermore, the balloon could not get to the anterior-midline part of vertebrae due to small extraversion angle, which might lead to the unsatisfactory reduction of the anterior vertebral body. To remedy these defects, some modifications were made on the unilateral trans-pedicle approach: First, the puncture point was translocated at least 5mm laterally from that of transpedicular approach. Second, the extraversion angle should be modified to about 40° so that the trocar could cross the outer wall of the pedicle , and eventually graze the inner wall when going into vertebral body. Through this approach, the balloon could dilatate in midline area and reduce the compressed vertebra symmetrically. Meanwhile, bone cement could be injected at this area and prone to be distributed bilaterally. Yan et al. [23] compared the outcomes between transverse process-pedicular PKP and bilateral pedicular PKP. Those researchers found that transverse process-pedicular PKP provided a similar symptom relief as control group but had significantly lower surgery duration and less radiation exposure dose. Another study by the same research team [24] showed that patients in the transverse process-pedicular PKP group had a better kyphosis correction due to the location of the balloon and bone cement injection than bilateral transpedicular PKP. Wang et al. [25] pointed out that the unilateral transverse process-pedicular PKP group had a similar clinical outcome as the conventional group, but a wider and symmetrical cement distribution, which contributes to in a better stress distribution. In previous literatures, although a few comparative studies between two approached of unilateral PKP were performed, comparison study among the three different surgical approaches of PKP was rarely reported. Therefore, in order to have a better understanding in this field, this study reviewed the clinical date of cases receiving PKP to determine the advantages, disadvantages and characteristics of the three different surgical approaches.
The results of pair sample and independent sample t test in our study indicated that, under similar preoperative general conditions, all three approaches of PKP could relieve pain, correct kyphosis and restore their self-care ability immediately, and no significant difference was found amongst the three surgical techniques in terms of symptom and kyphosis improvement rate. This finding illustrated that cement injection could restore the stability of fractured vertebrae in each of three approaches. Compared with empirical studies, we also found shorter surgical time in unilateral transpedicular PKP group. Although no significant difference of surgical time was found between the remaining two groups, the unilateral transverse process-pedicular PKP has a shorter operative time than bilateral transpedicular group. This phenomenon might be attributed to the low proficiency of the new surgical technique, which requires larger sample size and further studies to prove. When unilateral transverse process-pedicular PKP was performed, fluoroscopy was more frequently taken to confirm the location of trocar, but the surgical time should be drastically shortened with its proficiency developing. Meanwhile, a significantly higher cement leakage and lower cement volume were found in unilateral transpedicular PKP groups. In this group, the cement was injected from one side of vertebra and difficult to flow into the other side, leading to high proportion of only half cement distribution on fluoroscopy. In order to avoid this issue, cement should be injected quickly and more, which was also the main cause of cement leakage and shortened the surgical time indirectly. Besides, we found that part of patients in our study had an asymmetrical compression of vertebral height on the coronal section, and significantly higher coronal Cobb angle was found in unilateral transpedicular PKP group than two other groups, indicating this kind of PKP was prone to be chosen as the prior treatment for patients with asymmetrical vertebral compression in order to correct the spinal deformity at the same time. The results of our study showed that a significant improvement (4° or so) of coronal spinal deformity was noted after unilateral transpedicular PKP, implying this approach of PKP had a slight corrective ability of spinal deformity. In our opinion, it was very important for patients with degenerative scoliosis to avoid the acute progression of spinal deformity due to vertebral compression fracture. This finding was frequently ignored in preview studies.
Although positive results were found in our study, some shortcomings were also listed as follow. Firstly, small sample size, selection bias and limited follow-up duration might lead to the deviation of results. Secondly, unilateral transverse process-pedicular PKP was started from two years before, which generates proficiency of three approaches might interfere the result of this study. Last but not least, although both surgeons and measurers were skilled orthopedic surgeons, there were certain individual differences, which may lead to the deviation of the study results. Further work will be done to fill up the defect in future.