Introduction

Lumbar disc herniation (LDH) is a common condition with considerable impact on individual’s everyday life. In general, surgical treatment of disc herniation leads to fast leg pain-relief and a majority (75%) of the patients are satisfied with the surgical outcome (1). However, some patients experience continuous back pain and later undergo surgical procedures for their back pain problem, as fusion or disc prosthesis surgery. Our hypothesis was that preoperative characteristics on magnetic resonance imaging (MRI) representing a more degenerated disc would be associated with poorer outcome after surgery. The aim here therefore was to study if there was any association between preoperative IVD characteristics on MRI and back pain and back pain related disability 1-year postoperatively.

Methods

A total of 375 patients that had undergone lumbar disc herniation surgery at one hospital and been registered in the Swedish national spine surgery register (Swespine) 2012-2017 were identified. All patients with registered 1 year follow-up patient reported outcome measure (PROM) questionnaires (NRS, ODI, Global assessment (GA)) in Swespine and where the preop MRI could be localized were included in the study, n=218. Based on each of the 1-year reported PROM´s patients were divided into two groups: “unsuccessful” or “successful” PROM´s. On the MRI each herniated IVD was outlined on three consecutive T2-weighted midsagittal slices using semi-automated segmentation (Fig 1) and the mean signal intensity (MSI), standard deviation of signal intensity (SDSI) and SDSI/MSI of the midsagittal part of the IVD were calculated for the groups of patients. Further, to evaluate regional characteristics, each IVD was divided into 5 equally sized subregions, region of interest (ROI), based on the total midsagittal length of the IVD, ranging from 1 (anterior) to 5 (posterior). The different signal intensity measures between patients with “successful” versus “unsuccessful” surgical outcome were compared.

Results

The study population (n=218) had a mean age of 42.4 years at the time of surgery and consisted of 51.8% women. At 1-year follow-up they reported mean NRS back 2.9 (SD 2.7), and mean ODI 21.6 (SD 19.6). Descriptive statistics of the dichotomized groups based on PROM´s are described in Table 1. No statistically significant differences in signal intensity measures of the midsagittal part of the IVDs were found between the 1-year PROM outcome groups. The distribution of signal intensity measures (MSI, SDSI and SDSI/MSI) of the whole midsagittal part of the IVDs for the different groups are displayed in 2. For the subregion’s (ROI 1-5), no significant difference was found, except for SDSI in ROI 4 (p=0.045) comparing the NRS outcome groups.

Discusssion

This study could not prove any relationship between IVD characteristics and surgical outcome for most of the measurements. The only association found was for the SDSI of ROI 4 in the NRS back outcome group, where a small, but significant difference was detected. Based on the findings, the hypothesis could not be proven, implying that the disc characteristics measured here are not detailed enough and/or that experiencing continuous back pain after LDH surgery are multifactorial.