Spinal epidural abscess is well known to be a devastating condition and is associated with high rates of neurological disability, ranging from 15–27%, and mortality, ranging from 5–16% [7, 8, 20]. Surgical procedures such as decompression are of paramount importance for obtaining a good clinical outcome and a satisfactory recovery. However, robust evidence comparing surgical strategies such as decompression only versus decompression with instrumentation in older patients remains scarce, especially in those aged ≥ 80 years, who are exceptional because of their poor baseline reserve.
Summary of findings
To our knowledge, this is the first systematic analysis describing surgical strategies in octogenarians with PVO and concomitant SEA. The current study examined the clinical vignette, the neurological condition, the surgical characteristics, and the clinical course of this scarce disease with the aim of assessing morbidity and mortality rates. Interestingly, we found no significant differences between patients undergoing decompression only and those undergoing decompression with instrumentation regarding the following parameters: comorbidities, infection levels as determined by laboratory parameters (CRP and leukocytes), and grade of disability as defined by the MS score. Notably, patients undergoing pedicle screw fixation and surgical decompression had substantially longer surgical time and duration of hospitalization but no longer ICU stays. Also of note, among patients who opted for a surgical decompression procedure, we observed a significant improvement in the infection parameters and motor function at discharge, while in the instrumentation group, only the reduction in CRP levels reached statistical significance.
Literature review
Kim et al. retrospectively studied 16 patients aged > 65 years with PVO and SEA. Almost all patients (87.5%) underwent surgical decompression only [15]. Three patients were older than 80 years. Conversely, in the present study, we examined 35 patients older than 80 years, of whom 18 underwent surgical decompression only, and 17 underwent surgical decompression with instrumentation. Compared to Kim et al., who found a relatively low morbidity rate and a CCI as high as 4 in only 3 patients, our study cohort in both groups presented severe comorbidities with a mean CCI of 9.0 or greater [15]. Kim and colleagues did not report surgical duration, ICU stay, or hospital stay and observed a higher mortality rate after surgery (31.3%) [15]. In contrast, the in-hospital mortality in our group undergoing surgical decompression only was fundamentally lower (5.6%). The substantial discrepancy between the mortality rates might be attributable to the presence of delayed surgeries in the cited study, which were performed 40 days after the diagnosis. After completing the diagnostic work-up, our patients were operated on in less than 24 h with concurrent administration of IV antibiotics which might explain our low mortality rates even in older patients with multiple comorbidities.
Another retrospective analysis of 135 patients with PVO and SEA aged 18–88 years found that surgery led to significant improvements in neurological condition compared to medication alone [22]. It is important to highlight that about half of the cases were converted to surgical decompression and the other half to surgical decompression with instrumentation. However, the authors did not distinguish the outcomes of the two surgical approaches; thus, they broadly claimed that surgical treatment might be a critical pillar when treating patients with infections [22]. In another analysis of 60 patients with cervical SEAs and a mean age of 53 years, Alton et al. concluded that early surgery might be the key to the concurrent improvement of the infection and neurological statuses. Surprisingly, comorbidities did not predict treatment failure [1]. In agreement with these findings, Patel et al. compared medication versus surgical therapy in patients with PVO and SEA and advocated for early surgery based on an improvement of the MS by at least 3.4 points, also finding that diabetes mellitus, elevated CRP, and leukocytes were significant predictors of medical failure [18]. Based on the previous studies, the first-line therapy for such a devastating illness at our institution was surgery and not conservative management.
In a retrospective analysis of 40 patients with spinal abscess, Du et al. found that greater age (> 60 years) and the presence of comorbidities such as diabetes mellitus, respiratory, renal, or tumor diseases, and thrombocytopenia might significantly contribute to higher mortality rates evaluated 30 days post-surgery. They did not determine the surgical procedure [10]. Their overall 30-day mortality was 3.7%, comparable to those found by Darouiche et al. and Vakili et al. [7, 23]. Interestingly, the most frequently reported complications were septic shock, cardiac arrest, and pneumonia which were associated with mortality. In contrast, our mortality rates were substantially higher at discharge (8.7%) and at 90 days post-surgery (14.3%). This phenomenon might be attributable to our patient subset consisting only of patients older than 80 years and having many compounding factors (CCI > 6). Although each patient was postoperatively admitted to the ICU to prevent or treat postoperative complications, death was inevitable in some cases. Surprisingly, we found that older patients undergoing surgical decompression with instrumentation do not have a significantly greater mortality risk than surgical decompression alone. One explanation might be that the consistent use of spinal navigation in our surgical routine might be key for decreasing or even avoiding intraoperative complications and prolonged surgeries, thereby diminishing the rate of unanticipated postoperative events. However, it seems that surgical decompression alone can expedite the clinical recovery of this subset of patients compared to decompression with instrumentation.
Furthermore, it is important to highlight the occurrence of secondary instability after surgical spinal decompression, especially in older patients. Fox et al. suggested that radiologically confirmed secondary instability after spinal decompression is a common phenomenon that correlates with worse clinical outcomes or the necessity of additional fusion surgery [11]. Patients with preoperative instability signs, such as preoperative anterolisthesis or abnormal motion, on dynamic radiography, are at a higher risk of secondary instability after a spinal decompression procedure [16]. In our study, the mean follow-up was > 2 years, and no additional surgeries due to secondary instability were performed. We believe that advances in surgical techniques over the past decade, allowing adequate decompression with minimal disruption to the surrounding stabilizing factors, and the exclusion of patients with preoperative instability signs could have been reasons for the lack of revision surgeries in such a frail cohort. Nevertheless, a quite frequent phenomenon observed in older adults is the progressive degeneration of the spine as well as the spontaneous fusion, which might have contributed to higher rates of stability [4].
Strengths and limitations
The main strength of the current study is that we are the first to examine the outcomes of octogenarians undergoing surgery for SEA. However, this study has some limitations. First, we examined a relatively small cohort of patients. Nevertheless, since there is a lack of robust evidence of the clinical course of such a devastating disease in older individuals, we believe that our findings greatly clarify the clinical picture. Second, the minimum follow-up period of 12 months was relatively short; by gathering long-term data, other relevant findings not captured in the current study might have been revealed. Third, as this is a retrospective study, selection bias may have been present. Larger studies might be needed to elucidate potential candidates for non-operative management with antibiotic therapy only.