Introduction

Japan has one of the most aged populations. In 2006, Japan became the first superaged society, which is a society comprising > 21% of residents aged > 65 years1. The population of people aged > 65 years continues to increase, and it is predicted to continue to rise up to approximately 40% by 20652.

As a result of the population aging, the prevalence of degenerative spine disorders has increased, and the number of people who need surgical treatment is growing3,4. Advancements in diagnostic capabilities, surgical methods, techniques, spinal instruments, intraoperative navigation systems, and anesthesia have increased the number of spine surgeries. However, these elderly people are at risk of postoperative complications5,6,7,8. Postoperative complications can worsen surgical outcomes that may result in further disability and potentially increase morbidity6,8. Complications may also extend the period of hospital stay and lead to higher readmission rates9, which increases medical costs. Many previous studies have shown the risk factors of postoperative complications following spine surgery such as, older age, longer operative times, number of comorbidities, multilevel surgery, obesity, and so on8,9,10. Understanding the likelihood of and avoiding postoperative complications is critical for patient well-being.

Various scoring systems may predict the occurrence of postoperative complications in various surgical services. For example, the Surgical Apgar Score (SAS) is a scoring system based on intraoperative data is used to predict the occurrence of major complications or death within 30 days11. The SAS consists of three components: estimated blood loss (EBL), lowest mean arterial pressure (MAP), and lowest heart rate (HR), and is calculated by summing the score for each component (Table 1). The Controlling Nutritional Status (CONUT) was first reported as a screening tool to detect malnourished people which is based on serum albumin (ALB), total lymphocyte count (TLC), and total cholesterol (TC) values (Table 2)12. The CONUT score is useful for predicting the risk of postoperative complications13.

Table 1 The definition of Surgical Apgar Score.
Full size table
Table 2 The controlling nutritional status score.
Full size table

The SAS was developed for people who underwent general and vascular surgery, but the SAS may be applied in many other surgical services14. The CONUT score was first developed from hospitalized individuals from various medical departments. Many studies have applied the CONUT score in the field of general surgery and showed that it is useful in predicting the risk of postoperative complications13.

The usefulness of both tools has been shown in various medical specialties. However, to our knowledge, there are few reports specifically focused on using these tools in the field of spine surgery. The present study aimed to assess the utility of the SAS and the CONUT score in predicting the risk of postoperative complications following spine surgery. We hypothesize that both of the scoring systems can be useful in predicting the risk of postoperative complications in spine surgery, as they are for other medical specialties.

Material and methods

Study design

This study is a retrospective, single-center, observational study. We included data from people who underwent spine surgery in our institute between April 2013 and March 2020 in this study. We excluded the people who met the exclusion criteria: (1) People aged ≤ 19 years, (2) people with minor surgeries (e.g., debridement, percutaneous vertebroplasty, or biopsy), and (3) people with insufficient anesthetic records or laboratory data. The present study was performed in accordance with the ethical standards of Declaration of Helsinki and its later amendments. The study protocol was approved by the ethics committee of the University of Tsukuba Hospital. Written informed consent was obtained from all of the participants included in this study.

Data collection

For the background clinical information, age, sex, body mass index (BMI), comorbidities (diabetes mellitus [DM], hypertension, coronary disease, use of antiplatelet or anticoagulant agents), American Society of Anesthesiologists Physical Status Classification (ASA-PS), preoperative hemoglobin (Hb) value, and CONUT score were collected. The CONUT score was calculated based on the preoperative laboratory examination closest from the day of admission. Values of serum albumin (ALB), total lymphocyte count (TLC), and total cholesterol (TC) were used using the same method as the primary report of the CONUT score12. According to the scores, the participants were divided into four groups: normal (total score 0–1), light (total score 2–4), moderate (total score 5–8), and severe (total score 9–12). Based on our hospital’s rules, the laboratory examination was done within three months before the operation.

For the surgical information, surgical time, approach used in the operation, use of instruments, number of operation levels (> 5 or not), and SAS were collected. The SAS was calculated with the data collected from digital intraoperative anesthetic records. Information on intraoperative circulation dynamics which are the lowest MAP, EBL, and the lowest HR, was used. We calculated the SAS following the same method as the primary report11.

We also investigated the occurrence of major complications and the duration of postoperative hospital stay. Major complications were defined as events as follows: (1) acute renal failure, (2) bleeding requiring ≥ 4 units of red cell blood transfusion within 72 h postoperatively, (3) cardiac arrest, (4) deep venous thrombosis (DVT), (5) sepsis, (6) coronary artery disease, (7) unplanned intubation or ventilator use for over 48 h or longer, (8) pneumonia, (9) pulmonary embolism (PE), (10) severe delirium, (11) wound disruption, and (12) stroke11.

People who had any complications within 30 postoperative days were classified into the major complication group. We compared the background clinical information, surgical factors, and the length of hospital stay between the two groups that major complications occurred or not.

Statistical analysis

All of the statistical analyses were conducted using IBM SPSS Statistics (version 29, IBM, Armonk, NY). We used the Mann − Whitney U test to compare the average of the continuous variables (such as age, BMI, and surgical time), and the chi-square test to compare the categorical variables (such as ASA-PS, SAS, and CONUT). The continuous values are shown as the mean ± standard deviation (SD). We also used logistic regression analyses to obtain the independent risk factors for postoperative complications, and receiver operating characteristic (ROC) curves to determine the cutoff values. The variables with P < 0.1 in the univariate analysis were included in the logistic regression model. P < 0.05 was considered significant.

Results

Background information of the participants

There were 770 people who met the criteria for their data to be included. Of those, 111 had inadequate data or were aged < 19 years, so they were excluded. Finally, data from 659 people were included. The background clinical information of the participants whose data were included is shown in Table 3, and the diagnosis of the participants is shown in Table 4. Of the 659 people whose data were included, major postoperative complications occurred in 117 people (17.8%), and the details of the complications are listed in Table 5.

Table 3 The background information of the participants.
Full size table
Table 4 The diagnosis of the participants.
Full size table
Table 5 The detail of postoperative major complication.
Full size table

Comparison between the groups with and without major postoperative complications

Table 6 shows the results of univariate analyses. In the group that had major postoperative complications, there were significantly more people with DM (29.9% vs. 18.7%, P = 0.006), coronary disease (9.4% vs. 4.1%, P = 0.016), and antiplatelet therapy (18.8% vs. 10.3%, P = 0.010), and lower preoperative Hb values (12.8 ± 2.0 g/dL vs. 13.7 ± 1.8 g/dL, P < 0.001). There were more people with ASA-PS 3 or 4 (62.4% vs. 37.6%, P < 0.001) and their CONUT score was higher (2.25 ± 2.34 vs. 1.18 ± 1.39). Additionally, the average operation time was longer (430 ± 184 min. vs. 292 ± 143 min., P < 0.001), there were more operations using instruments (84.6% vs. 62.1%, P = 0.002), more multi-level surgeries (44.4% vs. 20.2%, P < 0.001), more anterior–posterior combined surgeries (13.7% vs. 5.5%, P = 0.002), and their average SAS was lower (5.0 ± 1.8 vs. 6.7 ± 1.5, P < 0.001). The length of postoperative hospital stay was significantly longer in the group tin which major postoperative major complications occurred (40 ± 20 days vs. 22 ± 14 days, P < 0.001).

Table 6 Comparison between groups with and without postoperative compliacations.
Full size table

The logistic regression analyses revealed that a history of DM (odds ratio [OR]: 1.81; 95% confidence interval [CI]: 1.04 – 3.13; P = 0.035), a history of coronary disease (OR: 3.33; 95% CI: 1.35 – 8.17; P = 0.009), ASA-PS (OR: 1.71; 95% CI: 1.03 – 2.83; P = 0.025), use of instruments (OR: 2.07; 95% CI: 1.09 – 3.94; P = 0.026), higher CONUT score (OR: 1.34; 95% CI: 1.15 – 1.52; P < 0.001), longer operation time (OR: 1.20; 95% CI: 1.13 – 1.34; P < 0.001), and lower SAS (OR: 0.59; 95% CI: 0.50 – 0.69; P < 0.001) were the independent risk factors for major postoperative complications after spine surgery (Table 7). The cutoff values of the CONUT score and the SAS were 1.5 points (area under the curve [AUC]: 0.646; sensitivity = 54.7%, specificity = 69.0%) and 5.5 points (AUC: 0.777; sensitivity = 80.4%, specificity = 65.8%), respectively (Fig. 1A, B).

Table 7 The result of multivariate analyses.
Full size table
Fig. 1
figure 1

The receiver operating characteristic curve indicates the cutoff for the scores in people with major complications. (A) The cutoff value for the CONUT score was 1.5 (area under the curve: 0.646; sensitivity = 54.7%; specificity = 69.4%). (B) The cutoff value for the SAS was 5.5 (area under the curve: 0.777; sensitivity = 80.4%; specificity = 65.8%).

Full size image

Discussion

We found that the SAS and the CONUT score can predict the risk of major postoperative complications following spinal surgery.

Although the SAS has been shown to be useful in predicting the occurrence of postoperative complications and 30-day mortality in various surgical services14,15, there have been only a few reports validating the use of the SAS, specifically in the field of spinal surgery. There are both reports showing both that the SAS was a predictor of postoperative complications and that the SAS was not a significant predictor. The usefulness was shown in a study in which the SAS was applied to people undergoing fusion surgery for lumbar degenerative diseases16 and cervical spine surgery17, but in people who underwent surgery for spinal metastasis, the SAS was not a significant predictor of postoperative complications18. The SAS was not useful in people with spinal metastasis because the occurrence of postoperative complications is likely multifactorial, and the condition of the people with cancer differs from those without cancer18. In the present study, the cohort included people with various diagnoses, unlike different from the previous reports, and suggests a relationship between poor intraoperative circulation dynamics and the incidence of postoperative complications following spinal surgery.

The relationship between preoperative malnutrition and postoperative adverse events after spinal surgery has been reported. According to those reports, a preoperative value under 50 in the prognostic nutritional status (PNI) was shown to be a risk for delirium19 and medical complications20 after adult spinal deformity (ASD) surgery, and surgical site infection following spine surgery21. By contrast, to our knowledge, only a few studies have validated the CONUT score in the field of spine surgery. One study reported that the CONUT score can be a significant predictor of major complications after cervical spine surgery17, and another showed that a higher CONUT score is associated with adverse outcomes after surgery for spinal tuberculosis22. Here, we showed that the CONUT score can predict the risk of major postoperative complications after spinal surgery, despite of the mixed cohort of participants whose data were included.

Here we show the possibility of reducing the risk of postoperative complications by assessing the CONUT score preoperatively and the SAS score postoperatively. In the present study, the period of postoperative hospitalization was longer in those in whom major postoperative complications occurred. We highlight that postoperative complications worsen surgical outcomes and extend hospital stays, increasing medical costs; therefore, preventing postoperative complications is essential. By detecting malnourished people before surgery, we can make the surgical plan less invasive to lessen the impact on circulation dynamics, or nutrition intervention can be considered for those who do not require urgent treatment. Preoperative nutrition intervention for malnourished people can reduce wound healing problems in elective lumbar spine surgery23 and postoperative medical complications in ASD surgery24, even though the objective nutrition status does not change. More intensive postoperative care should be considered in those with low SAS scores to avoid postoperative complications.

There are some limitations in our study. First, this is a retrospective single center study. There might be a selection bias in the people included in this study. Second, the exclusion criteria may have had the potential to impact the outcome of this study. People with insufficient anesthetic records and laboratory data were excluded, which may lead to the possibility that participants relevant to the research outcomes may have been excluded. A multicenter prospective study would be desirable to eliminate these biases and examine this study's validity. Finally, the postoperative nutrition status was not evaluated. The change in the nutritional status between pre and post-operation may be related to the occurrence of postoperative complications. This should be explored in a further study.

Conclusion

The present study showed that the SAS and the CONUT score can predict the risk of postoperative major complications after spine surgery. This study indicates that intraoperative poor circulation dynamics and preoperative malnutrition are related to postoperative complications. Both of the scores are easily calculated and universal. We recommend that both scores be used routinely to evaluate those who undergo spine surgery and determine which people are at risk of postoperative complications. Those who are detected should be managed carefully after spinal surgery.