Outcomes Reported After Surgery for Cauda Equina Syndrome

Supplemental Digital Content is available in the text

C auda equina syndrome (CES) is mainly caused by compression of the lumbosacral nerve roots below the conus medullaris. Clinically, symptoms and signs include low-back pain, saddle anesthesia, unilateral or bilateral sciatica, and motor weakness of the lower extremities with bladder and bowel dysfunction. 1,2 However, CES is a clinical-radiological diagnosis as clinical signs are not particularly sensitive to a CES diagnosis. 3,4 A lumbosacral magnetic resonance imaging (MRI) is required for diagnosis. Gleave and McFarlane 5 stressed the importance of categorizing CES into CES incomplete (CESI) and CES complete with urinary retention (CESR) (Figure 1). It is deemed a surgical emergency and there have been numerous publications and debates relating to the ideal timing for

LITERATURE REVIEW
surgery. [6][7][8][9] It can result in permanent damage to nerve roots resulting in long lasting or permanent disabling symptoms. 2 There is no consultation with patients in the literature regarding importance of outcomes for CES. In addition, there is little known about the long-term outcomes, which was highlighted by Korse et al, 10 who independently decided to focus on outcomes of micturition, defecation, and sexual function. Bias in studies, lack of universal definitions, and incomplete follow-up data were seen in this systematic review.
The problems with not having a core outcome set (COS) includes: (i) Patients are not included so important outcomes to them may not be measured. This has been witnessed in other healthcare areas such as childhood asthma and esophageal cancer. 11 At present, there is no COS for CES, which is to the detriment of patients and health services. The aim of this systematic literature review is to inform the future development of a COS by identifying all documented outcomes for patients after surgery in CES, identify if they are defined, and to assess what variability there is. The systematic literature review is the first step to inform the development of a COS 14 for patients who have undergone surgery for CES to be used in research and in practice.

METHODS
This study has been registered as 824 on the COMET (Core Outcome Measures in Effectiveness Trials) website (http:// www.comet-initiative.org/studies/details/824). Table 1 lists the inclusion criteria applied to the search strategy.

Search Strategy
We searched Medline, Embase, and CINAHL Plus (Cumulative Index to Nursing and Allied Health Literature). The search strategy for each database is available in Appendix 1, http://links.lww.com/BRS/B337. Online trial registries included Clinical Trials.gov, EU clinical trials registry and the ISRCTN (International Standard Randomized Controlled Trials Number) registry. The trial registries were searched for any completed or ongoing trials in surgery for CES and no relevant studies were found. Only case reports and abstracts were excluded in the initial search term as we wanted studies with five or more patients. We only included studies published after January 1, 1990 to keep investigation (post-MRI era) and surgical management of CES in line with current medical practice. Citations were collated with Endnote X7 referencing program (Thomson Reuters, New York, NY) and duplicates removed.

Data Extraction
Titles and abstract were initially screened by NS to identify potential studies for inclusion, for which full text articles were obtained for further assessment. Approximately, 10% of included articles were randomly checked for suitability by clinical supervisors and any discussion regarding uncertainty of eligibility criteria applied to the search results was discussed with them (SC, MW, and TM). A data extraction form was used to collect data on study design and location, patient demographics, timing of operation, definition of CES, diagnosis, etiology, surgical procedure, follow-up duration, outcome terminology, outcome definition, and assessment tool.

Terminology
Below are the definitions for the main terms used in the analysis of this systematic literature review.
1. Core outcome domain-The overall category to which similar subdomains and outcomes are listed under. The outcome domains that we have used in this article have been linked to the high level set of outcome categories used for annotation of Cochrane reviews 17 (http://linkeddata.cochrane.org/linked-data-project/ metadata-and-vocabularies/outcomes) and through discussion with the COMET initiative team. These are listed in bold in Table 3. 2. Subdomain-A subcategory of a Core outcome domain to which similar outcomes are listed under. These are listed in normal script in Table 3.

3.
Outcome-An outcome documented in an article after a patient has had an operation for CES. For example, nervous system (core outcome domain)> bladder function (subdomain)> urinary incontinence (outcome). 4. Variations-Variations were also documented, which means the number of different terms used to define a core outcome domain or subdomain. An example of a variation is given in the superscript of Table 4.  Table 4.

RESULTS
A total of 1873 articles were identified by electronic database searches. The Preferred Reporting Items for Systematic Reviews and Meta Analyses (PRISMA) flowchart in Figure 2 shows the process during the systematic literature review. Following inclusion criteria in Table 1 resulted in 1838 articles plus the 35 studies from the online registry search giving a total of 1873 studies. Moreover, 10% of included studies were reviewed by a supervisor (MW and SC) to assess if inclusion criteria had been applied adequately and agreement was achieved after discussion amongst us. Uncertainty regarding eligibility of certain full text articles for inclusion were discussed with the clinical supervisory team (MW, SC, and TM) and settled leading to 61 included articles. Thirty-four articles were excluded after the full text was obtained and the reasons for this were given as in Figure 2.
Summary details, patient demographics, and how many studies they were reported in out of the 61 included studies are detailed in Table 2. Most studies (90.2%) were retrospective. CES was not defined in 20 studies (32.8%). Even in the articles where CES is defined there were many differing definitions. The most common definition was CESI and CESR as described in Figure 1.    A total of 737 outcomes were reported in the 61 included articles. 9,18-78 For ease of analysis in this study, these reported outcomes have been categorized to one of the 20 core outcome domains ( Table 3). The nervous system core outcome domain had 10 subdomains, and the physical functioning has two subdomains ( Table 3). The number of different variations in the description of outcomes can be seen in Table 4 linked to the outcome domains. Figure 3 shows the number of articles in which specific outcomes were reported. Bladder function, motor function, sensation, bowel function, leg pain, and lower-back pain were the most commonly reported in descending order. They are all within the nervous system core outcome domain. Moreover, for each outcome, the number of articles where it is defined and not defined is documented. Figure 3 also shows the number of articles where the reported outcome had an assessment tool or not. Table 4 shows the raw data for each outcome showing how many studies each outcome is reported in, the total number of outcomes, the number of variations in the description of the outcome if a definition is present or not in the reported studies, and the number of assessment tools for the reported outcome. Table 5 shows the various assessment tools used for each outcome.

DISCUSSION
This systematic review shows that there is significant heterogeneity in the outcomes measured for patients who have undergone surgery for CES with no consensus.
Most of the evidence regarding outcomes for CES patients after surgery is derived from level 4 evidence, namely, single centre retrospective cohort review studies. The average data collection period was over 8 years with a median number of 14 patients per study, which highlights the rare nature of the condition and difficulty in collecting meaningful data retrospectively. This feeling is also echoed by Todd and Dickson, 2016. 79 Since 1990, the number of publications analyzing outcomes after an operation for CES have increased with the most being produced in the last 5year period (43.5%). Median follow up was at 31 months reflecting the deficiency in the literature for any longterm outcomes.
The main investigation is MRI, which reflects the systematic literature review focusing on studies from 1990 onwards. Before this there may have been a reliance on myelography and CT to radiologically identify CES compression. The main etiology is disc herniation. There are no studies in the literature documenting the exact distribution of CES aetiology but the most common cause is believed to be because of disc herniation.
Poor definition of CES has been previously highlighted in a systematic review of the literature. 80 Twenty studies (32.8%) did not define this and of the 41 studies where a definition was present, there was significant heterogeneity in the definitions. The most common definition for CES in this review was CESI and CESR. 5 If a study fails to define CES then we are unsure of the condition to which the outcomes of the study belong to. Most common surgical method in studies was a laminectomy and discectomy as seen in Table 2 but there were other studies that predominantly performed surgery via a microdiscectomy. Laminectomy alone, or with instrumentation was also mentioned for CES patients. In fact, now there is an increase in the popularity of endoscopic lumbar discectomy procedure 45 that adds to the range of procedures available when dealing with CES secondary to disc herniation. There is no consensus in the literature as to a specific decompressive procedure to be used for CES secondary to compressive pathology. This is also another factor that may affect outcomes for these patients.
In total, there were 737 outcomes reported verbatim and categorized into 20 core outcome domains and 12 subdomains. Instead of the same term being used for each outcome, there exists 507 variations in terminology (Table 4). In addition, most of the outcomes in the included articles have no definition. Except bladder function, adverse events, need for intervention, and global quality of life, all other outcomes had ''no definition'' in the majority of the included articles (Figure 3). This highlights that there is significant heterogeneity in not only the outcome terminology used but the level to which it is defined in the literature. Except global quality of life, emotional functioning, role functioning, and social functioning, most outcomes did not have an assessment tool in most of the articles (Figure 3). Fourteen of the outcome domains/subdomains we categorized had multiple different assessment tools used for each of them as seen in Table 5. There is a lack of uniformity over which assessment tool is best suited for each outcome in the literature. If outcomes are being measured with different scales, scoring systems, and questionnaires then it would be difficult to synthesize these results for meaningful analyses.
There is significant heterogeneity of the outcomes for patients who have undergone an operation for CES, how they are defined and measured in the literature. Bladder

LIMITATIONS
The systematic literature review was carried out by the main author (NS). Uncertainties and discrepancies were discussed with the research team (PRW, TM, MW, SC, and AN). Only English language articles were included. It would have been beneficial to have another independent group conduct the search strategy and data extract independently and to compare the results achieved. Because of the limitation of resources this was not performed.

CONCLUSION
There is significant heterogeneity in outcomes reported for studies after surgery for CES patients and the methods by which they are measured. This indicates a clear need for the development of a COS and the results of this systematic literature will be combined with the results of outcomes sourced from CES patients in qualitative interviews. All outcomes will then be prioritized through a Delphi process and consensus meeting to develop a core list of outcomes determined to be of most importance by key stakeholders.

Key Points
For patients who have had an operation for CES there are inconsistencies in the outcomes reported, defined, and assessed between studies. Because of the heterogeneity of outcomes reported, defined, and assessed we are unable to synthesize the results for a meta-analysis. The outcomes have not been validated in the literature by key stakeholders as being important to them.
Supplemental digital content is available for this article. Direct URL citations appearing in the printed text are provided in the HTML and PDF version of this article on the journal's Web site (www.spinejournal.com).