| dc.description.abstract | Low back pain is one of the most common health problems globally, having significant impact on individuals, community, and health care system. Lumbar Spine Surgery (LSS) is usually considered a treatment of low back pain when conservative management fails. In the United States, there has been an increase in the prevalence of LSS, with a similar increase in surgery costs and related post-surgical care. Although LSS is often considered to be a more efficient treatment than the nonsurgical management, the improvements gained from LSS are not optimal, resulting in no change or even worsening of symptoms in some cases. Investigating and understanding variables associated with surgical outcomes would be cost effective and clinically significance. Limited studies have attempted to examine patient recovery in acute state of LSS. Specifically studies are lacking to identify predictors of length of hospital stay (LOS), and little is known about predictors of discharge placement (DP) and outcomes as early as 2 weeks post discharge after having LSS. Bridging the gap in knowledge related to predictors of short-term LSS outcomes is the goal of this work. Identifying these predictors and early measures of recovery may lead to better utilization of resources and improved patient care and efforts to optimize LSS outcomes, Chapter 2 sought to identify predictors of LOS using various potential surgical and non-surgical variables. We used structural equation modeling analysis to study the direct effect of three latent factors on LOS: presurgical, surgical, and postsurgical factors. The three latent factors were constructed from potential predictor variables (indicators) that had significant direct effect on their related factors. Results showed that higher age diminished prior level of function, needing assistive devices, and low hemoglobin level were significant indicators of presurgical factors and associated with longer LOS. Secondly, high illness severity, increased complications, and need for intensive care unit stay after surgery were significant indicators of surgical factors and associated with higher LOS. Finally, inpatient physical therapy assessment, including low sitting and standing balance score, higher dependency in bed mobility transfer and mobility, and less distance walked during physical therapy sessions, were significant indicators of postsurgical factors and associated with longer LOS. The model explained approximately 50% of the variation in LOS. Postsurgical factors constructed from physical therapy assessment measured the highest percentage of the variation in LOS, followed by surgical factors, and finally presurgical factors that individually explained minor percentage of variation in LOS. Prospective studies are needed to confirm these results, and should consider including standardized clinical testing, especially at baseline to improve the prediction accuracy. Given that discharge placement (DP) predictors has been studies after many surgeries and conditions including total knee, total hip replacement, stroke and brain injury, little is known regarding the predictors of DP following LSS. Chapter 3 sought to address this gap in knowledge. Results showed that younger age, longer distance walked during hospital stay, and shorter length of hospital stay predicted greater likelihood of being discharged to home. Further analysis suggested that those living along, have inferior level of function prior to their surgery, and required longer hospital stay are likely to need skilled assistance (i.e. home health care or outpatient services) after being discharged to home ,. Prospective studies with more potential variables should be conducted to confirm these results. Short-and long-term outcomes following LSS were studied extensively following LSS. However, to our knowledge no study has investigated surgery outcomes earlier than 6 weeks post-hospital discharge. Therefore, chapter 4 explored the changes in patients' clinical status at 2 weeks following hospital discharge, and predictors of patient- outcomes during this short follow-up period. Results revealed that patients had significant reduction in back pain intensity, leg pain intensity, and improvement in function. However, there was no significant reduction in somatic perception and change in the type of analgesics used. High somatic perception predicted higher back pain, poor function, and inferior quality of life. Longer symptom duration was associated with higher postoperative back pain, while diagnosis of spondylolisthesis and preoperative use of opioids predicted higher postoperative leg pain intensity. Having high functional level at baseline was associated with high functional level postoperatively. Experiencing higher back and leg pain intensity, having depression symptoms, smoking, and receiving worker's compensation were significant factors associated with negative patient-perception of surgery outcomes. The study showed that multiple variables should be considered when predicting short-term LSS outcomes. In summary this dissertation work presented that LSS is effective in management of patients' pain, and improving function and quality of life for short-term follow up. Multiple variables showed to predict LOS, DP, and surgery outcomes after 2 weeks of post discharge after LSS. These variables could be presurgical variables including sociodemographic variables, cognitive behavioral variables, presurgical clinical status, presurgical functional level, or surgical including severity of illness, complications, longer intensive care unit and total hospital LOS, and postsurgical which including physical therapy functional assessment measures. The new knowledge presented in this work is important in guiding patients' selection criteria, establishing realistic expectations from surgery, and designing strategies to optimize surgery outcomes. Prospective studies with larger sample are needed to fully understand determinant of LSS success. | |
