Pregnancy Related Low Back and Pelvic Pain: a surgical approach

More than half of all pregnant women experience low back and/or pelvic pain of whom one-third has severe complaints. In most cases the pelvic pain disap­pears within a few months after delivery, either spontaneously or after con­servative treatment. In a minority of patients the pain persists even after a multidisciplinary rehabilitation program and may cause severe disability. Some patients may even be wheelchair bound or bedridden. After failure of all con­servative treatment, surgical fixation of the pelvic ring may prove one of the last remaining options for those women. Internal fixation of the pelvic ring is commonly used in unstable pelvic fractures. For pregnancy related low back and pelvic pain (PLBP), surgical fixation of the symphysis and sacroiliac joints (SIJ) has only been described in a few case reports and small series. In this thesis the biomechanical properties of different fixation techniques of the pelvic ring are investigated. Main subject of this thesis is surgical fixation of the pubic symphysis and the sacroiliac joints in patients severely disabled by pregnancy related low back and pelvic pain. In literature different nomenclature, definitions and classifications are used for pregnancy related low back and pelvic pain. Therefore in chapter two the com­plaints, diagnostic tests and criteria for the diagnosis pregnancy related low back and pelvic pain are discussed to establish the diagnosis pregnancy re­lated low back and pelvic pain. The following criteria were the most constant for the diagnosis PLBP: pain in one or both sacroiliac joints which originated during pregnancy or directly after delivery. The pain may radiate into the legs and is frequently accompanied by pain in the symphyseal region and pain in the groins especially at adduction of the hips. Usually complaints increase dur-ing exercise, which may impair Activities of Daily Life (ADL), like walking, climb­ing stairs, lifting objects and turning over in bed. In severe cases, patients walk with short steps and a waddling gate. The best validated pain provocation tests are the posterior pelvic pain provocation test (PPPP) and the active straight leg raising test (ASLR). The aetiology and pathogenesis of pregnancy-related low back and pelvic pain (PLBP) are subject of debate. Hormonal influences (relaxine) and mechanical effects (pelvic instability, postural changes and trauma) have been proposed as causative factors. A physiological increase in the width and vertical mobility of the pubic symphysis during pregnancy and a decrease af-ter delivery is described. Some authors reported a relationship between the severity of the pelvic pain and the amount of separation and movement of the symphysis, but the strength of the correlation varied widely in literature. Fur­thermore, a review is given of the conservative and operative treatment op­tions and the different surgical fixation techniques of the pelvic ring. In chapter three to six the results of our in vitro studies into the biomechanical properties of sacroiliac screw fixation are described. In unstable pelvic fractures, sacroiliac screws are one of the most commonly used methods for internal fixa­tion of the posterior pelvic ring and have the advantage of percutaneous place­ment. In order to determine the stability of different configurations and combi­nations of sacroiliac screw fixations, cadaveric pelves were loaded in a stand­ardised way. Translation and rotation stiffness of the fixation and the load to failure were measured using a three-dimensional video system. In chapter three to five sacroiliac screw fixation was tested in a biomechanical model represent­ing partial and completely unstable pelvic fractures, because in this model the stability of the fixed fracture depended largely on the stiffness of the osteosyn­thesis, which allowed a better comparison of the different fixation techniques. In pregnancy-related low back and pelvic pain, the structural integrity of the pelvic ring is intact. Therefore we used an intact isolated sacroiliac joint as a model in chapter six. In chapter three we compared different configurations of sacroiliac screws in order to find the optimal number and positioning in the sacral vertebral body. In 12 embalmed human pelvises a Tile C pelvic fracture was created, consisting of a symphysiolysis and sacral fractures on both sides. After cutting the pubic symphysis, the left and right sacroiliac joint were loaded separately as baseline measurements. Each of the sacral fractures was fixed with one of the following methods: one sacroiliac screw in the first sacral vertebral body, two screws convergingly in the first sacral vertebral body or one screw in the first and one in the second sacral vertebral body. The pubic symphysis was not stabilised, to limit the influence of the anterior fixation on the comparison of the screw tech-niques.The stiffness of the intact posterior pelvic ring was superior to any screw technique. The techniques with two screws showed a significantly higher load to failure and rotation stiffness than the moethod with one single screw in the first sacral vertebral body. There were no differences between the two tech­niques utilising two screws. The addition of a second screw seems to prevent rotation and improves the load to failure. In chapter four we studied whether the stability of partially unstable pelvic frac­tures can be improved by combining plate fixation of the symphysis with a posterior sacroiliac screw. In 6 specimens a Tile B1 (open book) pelvic fracture was created, by cutting the pubic symphysis and the anterior sacroiliac liga­ments. The pelves were loaded intact and after fixation of the fracture to 300 N, avoiding failure levels, and subsequently up to 700 N. The results showed no significant difference between isolated plate fixation and combined plate and sacroiliac screw fixation in either absolute displacements of the symphysis or sacroiliac joints or the stiffness. In addition, movements and stiffness of the fix­ated pelves were similar to the intact situation. Load to failure was only reached in one of the six cases. In all other cases the fixation of the pelvis to the frame failed before failure of the fixation itself. In these cases a load of about 1000 N or more could be applied. This suggests that the fixation could withstand even higher forces. Generally this is well above the force exerted by the upper body under physiological conditions. The addition of a sacroiliac screw in a Tile B1 fracture does not give significant additional stability and we recommend iso­lated plate fixation in Tile B1 fractures. Chapter five describes the stiffness and strength of combined anterior and pos­terior fixation under dynamic loading conditions in order to see if stability can be maintained in completely unstable (Tile C1) pelvic fractures. In 12 pelvic specimens a symphysiolysis and sacral fracture were created. We compared the intact situation to anterior plate fixation combined with one or two sacro­iliac screws. Each pelvis was loaded 2000 times, with a maximum of 400N, in the intact situation and after fixation with one of the two techniques. Further­more the load to failure and the number of cycles before failure were deter­mined. Translation and rotation stiffness of the intact pelvis were superior to the fixated pelvis. No difference in stiffness was found between the techniques with one or two sacroiliac screws. However a significantly higher load to failure and significantly more loading cycles before failure could be achieved using two sacroiliac screws compared to one screw. A better grip of the screws was a significant predictor of longer endurance of the fixated pelvis during loading. In this study embalmed aged pelves could be loaded repeatedly with a force which equals the upper body weight in adults. The fact that the average trauma patient is younger, suggests that direct postoperative weight bearing could be possible if these results are confirmed in further research. In chapter six we investigated whether 1 or 2 sacroiliac screws supply addi­tional stiffness to the intact sacroiliac joint, in order to make an estimation of the biomechanical properties of surgical stabilisation of the sacroiliac joint in PLBP patients. In 12 hemipelves baseline measurements of the intact sacroiliac joint without fixation were obtained, after which all sacroiliac joints were fix­ated sequentially with one and with two sacroiliac screws. In 10 cycli each hemipelvis was loaded to a maximum of 400N. For the technique with two screws a significantly higher translation and rotation stiffness and less displace­ment of the sacroiliac joint were found compared to the baseline. The differ­ence between one screw and the non fixated sacroiliac joint situation was less marked, but still significant for the translation stiffness. The rotation stiffness however showed no difference between one sacroiliac screw and the baseline. No significant difference could be found between the two screw techniques. In chapter seven we report on the functional outcome of internal fixation of the pelvic ring in a group of 58 patients suffering from severe pregnancy-related low back and pelvic pain in whom all conservative treatment has failed. Results were prospectively evaluated with the Majeed score, and endurance of walk-ing, sitting and standing. The surgical technique consisted of a symphysiodesis and bilateral percutaneous placement of two sacroiliac screws under fluoroscopic guidance. With a follow-up of on average 2.1 years, the difference between pre- and postoperative Majeed score indicated that an improvement of over 10 points was achieved in 69.8 % and 89.3% of the patients at 12 and 24 months respectively. Furthermore, a significant increase was found in walk-ing distance, endurance of sitting and standing and all Majeed score items (pain, work, sitting, sexual intercourse, walking aids, gait unaided and walking dis­tance). Improvement in mobility implied that of the 20 women who were wheel-chair-bound and of the eight women who were bedridden before the opera­tion, only four of the first group and four of the latter were using a wheelchair. The most important complications were irritation of nerve roots (8.6 %), non union of the symphysis (15.5 %), failure of the symphyseal plate (3.4 %) and pulmonary embolism (1.7 %). In this preliminary study surgical fixation of the pelvic ring yielded satisfactory results in severe PLBP patients in terms of pain relief, and improvement in ADL functions, although these results should be confirmed in a randomised clinical trial. Malpositioning of sacroiliac screws may lead to serious neurological complica­tions due to intrusion of the screws in the sacral foramina or vertebral canal. In chapter eight the safety of sacroiliac screw positioning using peroperative inlet and outlet fluoroscopy is assessed.We compared the correlation between screw position on peroperative fluoroscopy, postoperative radiographs and postop­erative CT scan. The radiographs, CT scan and charts from 88 patients, in whom the posterior pelvic ring was stabilised for several indications, were reviewed retrospectively. Seven of the 88 patients had neurological complaints and were reoperated. All complaints resolved completely and no permanent neurologi­cal damage occurred. Positioning both sacroiliac screws in the first vertebral body had a significantly lower rate of neurological complaints compared to positioning the lower screw in the second vertebral body. Malpositioning on CT scan correlated most accurately with neurological complaints, while no cor­relation between peroperative position and neurological deficit was found. 285 screws were reviewed and, depending on the type of imaging (X-ray or CT scan) 2.1% to 6.8% of the screw showed malpositioning. In several cases the malpositioned screws did not cause any complaints. Postoperative radiographs did not show to have any additional value above peroperative radiographs. In conclusion, percutaneous sacroiliac screws can be positioned safely, in experi­enced hands, without permanent neurological injury. In chapter nine the histological findings of the symphysis pubis of patients with severe pregnancy related low back and pelvic pain are analysed. Traumatic and degenerative changes of the pubic symphysis during and after pregnancy have been described in a few autopsy studies from the first half of the twentieth century, when mortality during pregnancy and labour was not exceptional. The relation of these changes with pelvic pain has never been described. A group of 15 women, who were seriously disabled by pregnancy related low back and pelvic pain, underwent internal fixation of the pubic symphysis and sacroiliac joints after all other conservative treatment had failed. These were the first patients of the study described in chapter seven. The histological changes of the symphyseal specimens removed during surgical fixation of these patients were compared to the symphysis of five healthy women. Vascular pro­liferation, callus formation, rupture of fibres, disturbance in the orientation of the fibres, and deposition of fibrinous material were seen in patients. A signifi­cant difference between the patients and the control group was found for rup­ture of fibres, and disturbance in the orientation of the fibres. No significant correlations could be detected between any of the pre- and postoperative outcome measures and individual or total histological characteristics. In con­clusion, degenerative changes of the symphysis pubis were found more often in patients with severe pregnancy related low back and pelvic pain than in control women. Finally, recommendations for further research are given in chapter ten. For a highly selected group of women, severely disabled by pregnancy related low back and pelvic pain, the results of surgical fixation of the pelvic ring seem promising in terms of pain relief and increased walking distance. However, these results should be confirmed in a randomised controlled trial. Furthermore, tests with prolonged dynamic loading and clinical studies are required to imple­ment the results of the biomechanical investigations in the postoperative weight bearing regimen for patients with unstable pelvic fractures. In the fu­ture, CT guided navigation and robotic insertion may result in a lower risk of peroperative neurological damage due to more accurate positioning of sacro­iliac screws.