Substantial attention has been paid to the effects of compliance with ERAS and the reduction of LOS[17–19]. To our knowledge, there are few studies focusing on the risk factors associated with prolonged LOS in short-level lumbar fusion surgery despite the implementation of ERAS protocol. In our study, we found that age \(\ge\)75 years, female sex, ASA \(\ge\)3, operation time, anesthesia time, delayed postoperative ambulation, prolonged removal urinary catheter, Clavien-Dindo grade >1 and allogeneic transfusion were associated with prolonged LOS. Binary logistic regression revealed that age female sex and severer postoperative complications were independently related to prolonged LOS in short-level lumbar fusion surgery, following a previous study to some extent[20, 21]. Regarding other risk factors, a rational interpretation for such an outcome was the distribution differences in the number of fusion levels. There were 51 (27.4%) patients with two fusion segments in Group 1 and 51 (70.8%) patients in Group 2, leading to different operation and anesthesia times and the number of patients undergoing allogeneic transfusion.
In a retrospective cohort study using the National Inpatient Sample database, Li et al showed elderly patients with comorbidities are at a higher risk for complications after lumbar spine surgery[20]. Analogously, in a retrospective analysis of 585 patients who underwent lumbar spine surgery, Kanaan et al found male patients walked significantly longer distance than female patients (β = 0.272 [95% CI (0.112–0.432)], p = 0.001), although clinical insignificant[22]. In our study, we found that age older than 75 years and female were independently related to prolonged LOS despite ERAS, which reminds us the importance of age and gender on postoperative complications, and more attention should be paid to these patients in perioperative management to prevent postoperative complications.
The increasing proportion of the aging population is generally associated with comorbidities and the postoperative complication rate following lumbar fusion surgery[23, 24]. ERAS results in shorter LOS and fewer postoperative complications than conventional perioperative care[5, 10, 25, 26]. Although ERAS has widely applied in other surgical fields, its implementation in spine surgery, though promising, is still in the early stages[7]. In a retrospective propensity score matching, D’Astorg et al revealed that the ERAS protocol decreased the hospital LOS (2.6 versus 4.4 days, p < 0.0001), while postoperative complications were similar in two groups[11]. In current study, according to Clavien-Dindo complication grade, we found patients with delayed LOS had severer postoperative complications (p < 0.001, Table 3), and Clavien-Dindo grade > 1 was independently correlated with prolonged LOS (Table 5). What’ more, it is worth noting that despite meticulous preoperative nutritional care and early oral feeding, there were 123 patients with postoperative hypoproteinemia (Table 2). According to previous studies, preoperative nutritional status is critical for recovery after major surgery correlated with postoperative complications and prolonged LOS[27, 28], therefore, it is essential to take additional measures to prevent severe postoperative hypoproteinemia.
Patients must actively participate in ERAS programs even though there may be reduced compliance with specific ERAS items[7]. In our study, the items with lower compliance were ambulation on POD 1 (63.9%) and removal of bladder catheter on POD 1 (70.9%). Previously, because of the reduction of bladder function and the longer functional recovery of spine surgery, very few studies explicitly illustrated the time of ambulation and removal of a bladder catheter and only stressed the concept of early ambulation and early removal of the bladder catheter[29, 30]. Given the adverse events associated with prolonged bed rest and the importance of early ambulation in reducing LOS[29, 31], attention should also be paid to adverse events related to prolonged urinary drainage (urinary tract infections and surgical site infections)[32, 33]. In a before and after cohort study on elderly patients who underwent elective spine and peripheral nerve surgery, Ifrach et al[34] showed improved mobilization and ambulation on POD 0 in ERAS group, and proved the safety and feasibility of early mobilization and ambulation in this population. In the present study, patients were explicitly requested to ambulate and remove the bladder catheter on POD 1 unless there were contraindications. Only one patient developed urinary retention (Table 2) in 132 patients (Table 3) who had the bladder catheter removed on POD 1, and there were no adverse events associated with early mobilization.
The present study has several limitations. First, this was a monocentric, retrospective study, and there were some offsetting and confounding factors. Second, the small sample size limits the robustness of our findings. Thirdly, due to the characteristic of spine surgery, the severe stress response and longer functional recovery, and the feature of the elderly population, patients with more than two fusion levels were excluded. This decision may have introduced selection bias. Finally, the imbalance of distribution of fusion levels between two groups may be the results of current study, hence, studies with comparable fusion levels are needed to further explore the risk factors associated with prolonged LOS.