Effects of early surgery for cervical fracture dislocation on 30-day mortality using the Japanese Diagnosis Procedure Combination database
Article information
Abstract
Study Design
Retrospective study of data abstracted from the Diagnosis Procedure Combination (DPC) database.
Purpose
This study aimed to investigate the effects of surgery in the early phase.
Overview of Literature
The optimal timing of surgery for cervical fracture dislocation (CFD) remains unclear because only a few clinical studies with approximately 100 patients have been published.
Methods
This study included 4,653 adult patients with a definitive diagnosis of CFD from the DPC database. The database contains nationwide inpatient data collected from >1,000 acute care hospitals in Japan. The DPC database contains information regarding hospitalization, such as diagnosis, treatment, medical history, complications, and hospitalization outcomes. This study identified 460 pairs of patients after one-to-one propensity-score matching (PSM). Treatment outcomes were compared between patients who underwent surgery for CFD within 72 hours (early group) and later (delayed group) after admission. The main outcomes included 30-day mortality, in-hospital death, and major complications. The secondary outcomes were improvement in the Barthel index, length of hospital stay, and discharged home rate.
Results
After adjusting for PSM, the early group had a significantly higher 30-day mortality rates than the delayed group (3.0% vs. 0.4%, p=0.006). In the multivariate logistic regression analysis after PSM, the early group was associated with an increased risk of 30-day mortality (odds ratio, 8.05; 95% confidence interval, 2.15–5.26; p=0.007).
Conclusions
This study indicated that early surgery for CFD resulted in increased 30-day mortality.
Introduction
Cervical fracture dislocation (CFD) has a relatively poor prognosis because of severe instability and spinal cord injury (SCI) [1]. Treatment and stabilization in the early phase are thought to be essential. Although no clear consensus has been reached on the optimal timing for CFD treatment, surgery is recommended for CFD with SCI [2]. No nationwide studies have examined the outcomes based on the timing of surgery for CFD because of the extremely limited number of cases. Therefore, this study aimed to investigate the effects of early surgery to determine whether the period to surgery affects treatment outcomes using the Diagnosis Procedure Combination (DPC) database, which is a national database containing inpatient data collected from >1,000 acute care hospitals in Japan.
Materials and Methods
Data source
The DPC database is a national database containing administrative claims from >1,000 hospitals and includes approximately 50% of all acute care inpatients in Japan [3]. This database contains information regarding hospitalization, such as diagnosis, medical treatment, medical history, complications, and hospitalization outcomes. This database specifies the International Classification of Diseases, 10th Revision (ICD-10) codes for primary and secondary diagnoses, preexisting comorbidities on admission, and complications after admission.
For this study, the need for informed consent was waived because of data anonymity. The Institutional Review Board at the University of Tokyo Medical and Dental University approved this study (approval no., M2000-788-29).
Study participants
This study included inpatients with a definitive diagnosis of cervical fracture (ICD-10 codes: S122 and S129) who had main or comorbid diseases including CFD with or without SCI between 2010 and 2021. A total of 4,653 patients were identified. The exclusion criteria were as follows: (1) aged <20 years, (2) not admitted for the first time for CFD, (3) died within 24 hours after admission regardless of the cause of death, (4) not admitted on an emergency basis, and (5) multiple trauma that might affect clinical outcomes such as severe head injury or pelvic fracture (Fig. 1). This study included 2,750 patients, with 2,083 male and 667 female patients. The mean age was 66.6±16.2 years, and age distribution revealed a peak at approximately 60–80 years (Fig. 2). Among these 2,750 patients, 1,619 underwent surgery, excluding immobilization with a halo-vest. These patients were divided into two groups according to the treatment timing: patients who underwent surgery for CFD within 72 hours after admission (early group) and >72 hours after admission (delayed group). The cutoff value of early surgery timing was set to 72 hours based on previous findings [2,4]. Finally, 928 and 691 patients were included in the early and delayed groups, respectively. This study identified 460 pairs of patients after one-to-one propensity-score matching (PSM) (standardized mean difference of <0.1). Table 1 shows the patient demographics of these two groups before and after PSM.
Flow chart of this study.
Age distribution of the patients diagnosed with cervical fracture dislocation.
Clinical characteristics of the two groups before and after 1:1 PSM
Data extraction
Clinical variables on admission were identified from the DPC database including age, sex, Japan coma scale, Barthel index (BI), and preexisting comorbid conditions, such as the Charlson comorbidity index (CCI), anticoagulant medications, antiplatelet therapy, or steroid pulse therapy. The CCI score summing the given scores for comorbidities were used to categorize the patients into four groups [5]: none (0 points), low (1–2 points), intermediate (3–4 points), and high (4 points).
The main clinical outcomes were as follows: within 30-day mortality, in-hospital death, and major complications after admission, such as pulmonary embolism or deep vein thrombosis, respiratory complications, cardiac events, cerebral infarction, gastrointestinal bleeding or ulcer, sepsis, urinary tract infection, and delirium. The secondary outcomes were as follows: improvement in BI (difference between BI scores at admission and discharge), length of hospital stay, and discharged home rate (rate of patients discharged directly to home).
Statistical analysis
R ver. 4.1.2 (The R Foundation for Statistical Computing, Vienna, Austria) was used for all statistical analyses, with Student t-test or the Mann-Whitney U test for the comparison of continuous data, and the chi-square test or Fisher’s exact test for the comparison of categorical data between the groups. One-to-one PSM in the early and delayed groups was performed. Multivariable logistic regression analysis for 30-day mortality was conducted in the PSM-matched cohort. A two-tailed significance level of p-values of <0.05 was used in all analyses. Data are given as the mean±standard deviation.
Results
The primary and secondary outcomes after adjusting for PSM are shown in Table 2. The 30-day mortality rate was significantly higher in the early group than in the delayed group (3.0% versus 0.4%, p=0.006). The groups did not differ significantly in other main outcomes, including in-hospital death and major complications (p>0.05, respectively). Moreover, the early group had poorer outcomes than the delayed group in terms of improvement in BI (p=0.005) and discharged home rates (p=0.001). In the multivariable logistic regression analysis for 30-day mortality, early surgery was significantly correlated with higher 30-day mortality rates in the PSM-matched cohort (odds ratio, 8.05; 95% confidence interval, 2.15–5.26, p=0.007) (Table 3).
Primary and secondary outcomes of the two groups after 1:1 propensity score matching
Risk factors of 30-day mortality in multivariable logistic regression analysis
Discussion
This study demonstrated that the 30-day mortality rate was significantly higher in the early group even after adjusting for PSM. Moreover, the poorer outcomes in the early group were identified by the evaluation of improvement in BI and the discharged home rate. In addition, early surgery increased the 30-day mortality rate in the acute phase. The widespread belief that early stabilization for spinal fractures achieves better outcomes has influenced the timing of surgery [6]. When limited to CFD, the optimal timing from injury to surgery remains unclear. The time from injury to surgery of >3.8 days, a subaxial injury classification score of >7.5 points, and spinal cord compression of >55.8% are risk factors for poor functional prognosis of CFD with SCI [2]. The mortality rate could be reduced if the initial surgery for spinal fractures was performed within 72 hours after injury; however, this study did not focus exclusively on patients with CFD [4]. The appropriate timing of surgery for CFD has little evidence-based data. Thus, our results from the analysis of data from 1,619 patients provide unprecedentedly novel insights.
We referred to a previous study [4] to set the cutoff value of early surgery timing to 72 hours based on the findings of previous studies [2,4]. Internal fixation in the early phase can achieve rigid mechanical stabilization and anatomical reduction of fracture segments and helps decrease complications [7,8]. Conversely, assessment using the Frankel classification demonstrated no significant difference in final neurological outcomes between early (<72 hours) and delayed (>72 hours) surgery [8]. However, all patients underwent preoperative cranial traction [8], which possibly affects the neurological outcomes.
This study revealed that early surgery may have resulted in high 30-day mortality rates. The first possible hypothesis is that sympathetic hyperactivity may occur after injury in the acute phase as in traumatic brain injury [9]. Similar to previous findings, early surgery for spinal fractures leads to significantly higher mortality rates among patients with poor clinical conditions [10]. These outcomes are also caused by impaired vasomotor function, possibly leading to severe hemodynamic instability [10]. Therefore, the addition of surgical invasiveness to this sympathetic hyperactivity in the early post-traumatic phase may have caused excessive stress, which could lead to higher 30-day mortality rates.
Another hypothesis suggests that operative invasion worsens minor immune function deficiencies in the early period [6,11]. Patients with severe damage may also receive a “first hit” (initial injury) followed by a “second hit” (surgery [11], iatrogenic injury [12]) injury that may amplify the first hit, which further stimulates the already activated inflammatory cascade. This can result in increased 30-day mortality rates [6]. However, to prove these hypotheses, little supportive data are available; thus, further studies including randomized controlled trials (RCTs) are needed.
This study has several limitations. We tried our best to remove bias by PSM; however, limitations remain. First, specific clinical information that can exactly identify the level of CFD, severity of SCI (Frankel classification or American Spinal Injury Association impairment scale), presence of interlocking, and dislocation type are unavailable. In addition, the DPC database does not contain data about the cause of in-hospital death. Second, the DPC database shows only information during hospitalization; therefore, the exact time of injury is unknown. However, the dates of injury and admission are considered the same in most cases, so no large difference was presumed. Third, treatment strategies varied in terms of surgical strategies (conservative treatment, anterior/posterior/combined approaches) or timing. Individual surgeon’s preferences and radiological findings, such as traumatic disc herniation, deformity, and malalignment, may have introduced bias in treatment selection. Despite these limitations, this study provides valuable and novel insights into the continuing debates about the optimal timing of surgery, particularly for CFD. Further clinical studies for CFD with more detailed data or RCT are warranted to clarify novel findings.
Conclusions
This nationwide study investigated the largest dataset recording the timing of surgery for CFD. Early surgery for CFD resulted in higher 30-day mortality rates. Further clinical studies for CFD are warranted to evaluate the optimal timing of surgery.
Key Points
Treatment outcomes were compared between patients who underwent surgery for ervical fracture dislocation (CFD) within 72 hours (early group) and later (delayed group) after admission.
This nationwide study indicated that early surgery for Diagnosis Procedure Combination resulted in increased 30-day mortality from the DPC database.
Further clinical studies for CFD are warranted to evaluate the optimal timing of surgery.
Notes
Conflict of Interest
No potential conflict of interest relevant to this article was reported.
Funding
This work was supported by a Grant-in-Aid for Research on Policy Planning and Evaluation from the Ministry of Health, Labour and Welfare, Japan (grant no., 22AA2003). The funder has played no role in the design, data collection, analysis, decision to publish, or preparation of the manuscript.
Author Contributions
Conceptualization: KD, NI, NO; data curation: KD, NI; formal analysis: NI; funding acquisition: KF; methodology: NI; project administration: AY; visualization: NO; writing–original draft: KD; writing–review & editing: JM; final approval of the manuscript: all authors.