Letter: Steroid Use Associated With Increased Odds of 30-Day Mortality in Surgical Patients With Metastatic Spinal Tumors in the Setting of Disseminated Disease

To the Editor: We read with great interest the clinical study by Hobbs et al, “Steroid Use Associated With Increased Odds of 30-Day Mortality In Surgical Patients With Metastatic Spinal Tumors in the Setting of Disseminated Disease”.1 However, we were surprised to find a critical flaw in the manuscript's study design, which puts the authors’ conclusion and overall “take-home” message in serious doubt. In their study, the authors used the prospectively collected American College of Surgeons National Quality Improvement Program (ACS-NSQIP) database in an attempt to identify patients who received preoperative steroids for metastatic spinal cord compression (MSCC) and found that steroid use was associated with a significantly higher risk of 30-d mortality.1 In fact, the authors’ conclusion suggests that administration of steroids upon initial presentation of motor weakness in patients with MSCC increases the risk of mortality. However, the ACS-NSQIP database does not include data on steroid administration for this purpose. The “steroid” variable in the ACS-NSQIP database is used to identify patients that “have required the regular administration of oral or parenteral corticosteroid medications or immunosuppressant medications, within the 30 d prior to the principal operative procedure or at the time the patient is being considered as a candidate for surgery, for a chronic medical condition (eg, COPD, asthma, rheumatologic disease, rheumatoid arthritis, inflammatory bowel disease)”.2 The analyzed data, therefore, encompasses patients who had been on long-term steroids, perhaps for months or even years, and not specifically patients who received steroids upon acute presentation with MSCC. Furthermore, it is possible that some of those medications were not even steroids but rather immunosuppressants for chronic conditions. The authors indicate that they were “quite surprised” by the findings.1 In fact, their findings would be a deviation from the current gold standard. A recent systematic review by expert spinal surgeons who routinely take care of patients with MSCC found that steroids do have an important role in this population in maintaining ambulatory status at 1 yr. Although steroids were found to be associated with gastrointestinal complications, no significant increase in mortality was reported in the meta-analysis.3 On the contrary, maintaining the ability to walk is critical in this patient population and can improve survival.3 Sorensen et al4 published almost 25 years ago a single-blinded randomized trial comparing high-dose dexamethasone (96 mg IV bolus followed by 24 mg every 6 h tapered over 10 d) plus radiation to no steroids plus radiation in patients with MSCC, finding significantly higher rates of ambulation at 1 yr and yet no increase in mortality in favor of steroids. Interestingly, the utilized doses were much higher than what most neurosurgeons would use nowadays or what is currently recommended (typically an initial dose of 10 mg followed by 4 mg every 6 h), and yet there was no increased risk of death.4 Hobbs et al1 dedicated a considerable part of their efforts to conduct exhaustive statistical analyses to further verify their results. This is commendable but ultimately irrelevant in view of the fundamental flaw in the collection of the raw data. Despite briefly acknowledging the limitations of their study, the authors do not take into account that this acknowledgment per se discredits their ultimate conclusion. The authors comment that chronic conditions were evenly distributed among cohorts and therefore “this information increases the likelihood that steroids were administered for the MSCC disease itself”.1 There are numerous variables not included in the ACS-NSQIP database that may have explained discrepancies in steroid use—rheumatoid arthritis, lupus, post-transplant patients, etc. As there is no way that the authors can be sure that steroids were administered for MSCC, the design of the study does not allow the conclusions that were made. In fact, it would be more accurate to say that the results of this study indicate that chronic steroid use in patients with MSCC was associated with a significantly higher risk of 30-d mortality. Multiple investigations using this database have actually shown that chronic steroids increase the risk of mortality after colectomy in ischemic colitis,5 bariatric surgery,6 and others.7 The distinction is essential as any other conclusion may have potentially detrimental effect on the current management of patients with MSCC and neurological deficits. The use of large national/multi-center databases for research in neurosurgery has exponentially increased in the last few years.8 While these databases do certainly play a role in current research and allow for analyses of very large patient populations, it is imperative to know both the capabilities and also the limitations in their design. The authors attempted to evaluate the effect of steroid administration for MSCC on the risk of 30-d mortality. However, based on the limitation imposed by the database they utilized, they instead most likely identified a group of patients that had been on steroids or even immunosuppressants for months/years before even presenting with neurological deficits from their tumors. Readers of this manuscript should acknowledge this important issue and not be discouraged to use steroids judiciously at recommended doses in patients with MSCC. Disclosures Dr Yassari has a consulting agreement with Stryker. The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.