There is not yet a high-evidence study comparing the efficacy and safety of early and late decompression in a subpopulation with different patient characteristics. However, in some studies, this was analyzed and reported at the level of suggestion [45]. In the subpopulation according to age, there is no study comparing the effectiveness and safety of early decompression and late decompression between younger and older people. However, there is a recent study comparing the clinical outcome of surgical management between younger and older patients with acute cord injury. According to Lau et al. [28], the complication rate and mortality rate of older patients (≥70 years) were 1.7 times and 10.8 times higher than those of younger patients (<70 years), respectively. However, there was no difference in discharge American Spinal Injury Association impairment scale (AIS) grade and AIS grade change between the two groups. In the subpopulation according to the degree of cord injury, patients with AIS B, C, and D injury at C2-L2 level before surgery had an additional 6.3 motor point motor recovery improvement with early (≤24 hours) decompression than with late (>24 hours) decompression. In patients with AIS A injury, there was no beneficial effect of early decompression on motor recovery. However, in patients with AIS A or B injury, early decompression resulted in a shorter hospital stay [17]. Also, as the AIS went from A to D at admission, when decompression was performed within 12 hours, the conversion rate from AIS grade at admission to better AIS grade at discharge was higher [39]. Moreover, ultra-early (≤12 hours) decompression showed more improvement in AIS than decompression after 12 hours, regardless of the degree of preoperative cord injury. In particular, it has been reported that even in patients with AIS A injury, if ultra-early decompression was performed, the AIS grade improved up to stage 1 or 2 after surgery [12]. However, although only for cervical cord injury, other studies have reported that timing of decompression (ultra-early, <12 hours; early, 12-24 hours; and late, >24 to 138.5 hours) does not affect AIS grade conversion, and intramedullary lesion length on preoperative magnetic resonance imaging is a major predictor of postoperative neurologic outcome [11]. In a subgroup analysis according to the use of preoperative steroids, early (≤24 hours) decompression in patients receiving preoperative steroid showed more improvement in pin prick score, light touch score, and total motor score than late (>24 hours) decompression. There was no significant difference in total motor score improvement between early and late decompression in patients who did not receive steroids before surgery [35].

As mentioned earlier, there have been many clinical studies on the effect of early decompression in SCI. However, it was difficult to standardize the time due to heterogeneity in the definition of “early” for each study, and there was a limit in achieving strong recommendation. Most of the studies that see the time threshold of the definition of “early” as 24 hours, this time window is most appropriate to define “early” [21]. Furthermore, A prospective, multicenter study, the Surgical Timing in Acute Spinal Cord Injury Study (STASCIS) in 2012, defined “early” as first 24 hours, and reported that decompression within 24 hours after SCI was associated with improved neurologic outcomes showing at least grade 2 AIS improvement at 6 months follow up [19,20]. Later, many studies reported the efficacy of early decompression (within 24 hours) in patients with SCI [21,42,44]. However, the timing of decompression for acute SCI remained controversial. In particular, in patients with central cord syndrome without instability, since spontaneous improvement may occur, decompression of the sensitive and fragile spinal cord due to injury may result in neurologic worsening. In 2017, AOSpine (Arbeitsgemeinschaft für Osteosynthesefragen spine) reviewed the efficacy, and safety of early (≤24 hours) versus late (>24 hours) decompression in patients with SCI [45]. The effects of early decompression on neurologic improvement in SCI patients varied according to the study and injury level. There was no difference at other levels, but at the cervical cord level, early decompression showed statistically and clinically significant improvement 6 months after decompression. In addition, there was an effect of 6-point improvement on the AIS in patients with AIS B, C, and D, but not in AIS A. In patients with acute central cord syndrome without instability, the total motor score improved at 6 and 12 months in early decompression, but there was no difference in AIS improvement between early and late decompression group. The rates of complications also did not differ between the two decompression groups [45]. Based on these contents, in the same year, AOSpine presented a clinical practice guideline for the timing of surgical decompression in patients with acute SCI [18]. They suggested that early decompression should be considered as a treatment option in adult patients with traumatic central cord syndrome and that early decompression should be offered at any level in adult patients with acute SCI. Previous animal study has reported that spinal cord ischemia caused by swelling and hemorrhage reaches peaks at 8 hours after injury [16]. Grossman et al. [23] suggested that if SCI patients visit the emergency room immediately after injury, most arrive within 4 hours, and that 8 hours to surgical decompression is a realistic goal. Based on these contents, an 8-hour time window was established. In previous randomized controlled trials [13], decompression within 8 hours in SCI patients showed better neurological results than more than 8 hours. In a study comparing the decompression within 8 hours versus 8-24 hours, neurologic outcome is better in decompression within 8 hours than 8-24 hours after injury [27]. More recently, many papers have reported the positive impact of decompression within 8 hours after injury [29,46,47]. However, in some studies, the definition of ‘within 8 hours after injury’ was measured as the time taken after arriving at the emergency room. Therefore, it is difficult to directly compare the results of remaining studies. Also, in the narrow time window of 8 hours, consumption of 1 or 2 hours is very important. In actual clinical practice, surgical intervention within 8 hours after injury is practically difficult due to availability of operating room, transport time to hospital, and lack of availability of medical team. There is also a study that reported negative results for early decompression. Tanaka et al. [40] analyzed the patients by dividing them into an early group (within 24 hours) and a late group (from 25 hours to 7 days). The length of intensive care unit stay, in-hospital mortality, and survival rate were not different between the two groups [40]. A meta-analysis study from pooling of individual patient data from four high-quality, prospective, multicenter acute SCI datasets, including data from December 1991 to March 2017 also concluded that there is a 36-hour time window for neurologic benefit and suggested a 36-hour time window for high probability of neurologic improvement. Subject of the exact timing of surgery for decompression after SCI is still very controversial [5,8,19,22,24]. In that sense, the Prospective, Observational European Multicenter (SCI-POEM) study, which compares the neurological outcomes of decompression within 12 hours and after 12 hours after injury, has recently completed enrollment of 300 people and will publish the results soon [41]. The results will be very helpful in finding the best cutoff point of timing of decompression after SCI.