Long-term outcomes following vehicle trauma related acute kidney injury requiring renal replacement therapy: a nationwide population study

Acute kidney injury (AKI) is a frequent complication of traumatic injury; however, long-term outcomes such as mortality and end-stage kidney disease (ESKD) have been rarely reported in this important patient population. We compared the long-term outcome of vehicle-traumatic and non-traumatic AKI requiring renal replacement therapy (AKI-RRT). This nationwide cohort study used data from the Taiwan National Health Insurance Research Database. Vehicle-trauma patients who were suffered from vehicle accidents developing AKI-RRT during hospitalization were identified, and matching non-traumatic AKI-RRT patients were identified between 2000 and 2010. The incidences of ESKD, 30-day, and long-term mortality were evaluated, and clinical and demographic associations with these outcomes were identified using Cox proportional hazards regression models. 546 vehicle-traumatic AKI-RRT patients, median age 47.6 years (interquartile range: 29.0–64.3) and 76.4% male, were identified. Compared to non-traumatic AKI-RRT, vehicle-traumatic AKI-RRT patients had longer length of stay in hospital [median (IQR):15 (5–34) days vs. 6 (3–11) days; p < 0.001). After propensity matching with non-traumatic AKI-RRT cases with similar demographic and clinical characteristics. Vehicle-traumatic AKI-RRT patients had lower rates of long-term mortality (adjusted hazard ratio (HR), 0.473; 95% CI, 0.392–0.571; p < 0.001), but similar rates of ESKD (HR, 1.166; 95% CI, 0.829–1.638; p = 0.377) and short-term risk of death (HR, 1.134; 95% CI, 0.894–1.438; p = 0.301) as non-traumatic AKI-RRT patients. In competing risk models that focused on ESKD, vehicle-traumatic AKI-RRT patients were associated with lower ESKD rates (HR, 0.552; 95% CI, 0.325–0.937; p = 0.028) than non-traumatic AKI-RRT patients. Despite severe injuries, vehicle-traumatic AKI-RRT patients had better long-term survival than non-traumatic AKI-RRT patients, but a similar risk of ESKD. Our results provide a better understanding of long-term outcomes after vehicle-traumatic AKI-RRT.


Scientific Reports
| (2020) 10:20572 | https://doi.org/10.1038/s41598-020-77556-3 www.nature.com/scientificreports/ suggested management of AKI in settings such as sepsis or nephrotoxin exposure, comparatively less data is available to guide clinicians on the clinical course of AKI after major trauma. Trauma is a leading cause of hospitalization worldwide, which mainly affects a young and previously healthy population both in the developed and developing world. The incidence of AKI among major traumatic injuries has been reported to be 6.0-36.8% 6,7 , and traumatic AKI has been associated with a mortality rate of 14.9-57.0% [7][8][9][10] . Overall around 5% of trauma patients admitted to the intensive care unit (ICU) require RRT 10 . In a recent systematic review the pooled incidence of AKI after trauma was 20.4% with an associated 3.6fold increase in the relative risk of death, commonly reported associations of AKI after trauma include older age, higher severity of injury, volume of blood transfusion, abdominal site of injury and presence of comorbid disease including diabetes mellitus 11 . Prior studies of traumatic AKI-RRT have focused mainly on short-term outcomes 12,13 and detailed comparisons of outcomes between traumatic-AKI and AKI of differing etiologies that have not been made.
Given the paucity of studies concerning traumatic AKI-RRT, we sought to describe its epidemiology and associated short and long-term outcomes in a large epidemiological database.

Results
Temporal change in in-hospital mortality. Among 123,470 hospitalized AKI-RRT patients in the past decade, we identified 546 vehicle-traumatic AKI-RRT patients who survived to index discharge (Fig. 1). The annual number of trauma cases decreased gradually; however, the proportions of vehicle-traumatic AKI-RRT patients tended to increase over time, especially in the elderly group (aged ≥ 65 years) (Fig. 2). In our cohort, young patients (aged ≤ 44 years) were less likely to be admitted because of trauma; however, they had the highest incidence of vehicle-traumatic AKI-RRT (Fig. 2) during hospitalization. On the other hand, soft tissue injury from trauma had the lowest incidence of AKI-RRT (Fig. 3).
Patient characteristics before and after propensity score matching. Compared to patients with other causes of AKI-RRT, vehicle-traumatic AKI-RRT patients were younger [median IQR: 47.6 (29.0-64.3) vs. 65.9 (53.7-74.5), p < 0.001) and more of them were male (77.1% vs. 54.2%, p < 0.001). Vehicle-traumatic AKI-RRT patients also had a lower rate of cerebrovascular disease, COPD, diabetes, liver disease, and advanced chronic kidney disease (CKD) which was defined as the presence of CKD coding and concomitant reimbursement coding of erythropoiesis-stimulating agents without copayment. More pulmonary and cardiac complications occurred in AKI-RRT patients than non-traumatic AKI-RRT patients during hospitalization. Among them, 517 vehicle-traumatic AKI-RRT patients were successfully matched with 1,551 non-traumatic AKI-RRT patients (Table 1).
To evaluate the potential etiology of AKI in our cohort (Supplemental Table 1), we collected the data of inhospital acute comorbidities, including hypovolemic shock, cardiogenic shock, and severe sepsis which may relate to AKI. In addition, we also collected the data of contrast exposure and nephrotoxicity medications exposure Outcomes in patients with AKI-RRT after discharge. After propensity score matching, the followup period was 2.98 years. The 30-day mortality was similar between two groups (18.4% vs. 15.9%, p = 0.207). Although the baseline comorbidities and acute medical problems before index hospitalization were similar in both groups after matching, vehicle-traumatic AKI-RRT patients had a longer length of stay in hospital [median IQR 15 (5-35) vs. 12 (5-28) days, p = 0.007). The long-term incidence of ESKD was similar between groups (10.6% vs. 13.4%, p = 0.832) ( Table 1). www.nature.com/scientificreports/ The risk of developing ESKD (HR, 1.166, p = 0.377) and 30-day mortality after discharge (HR, 1.134, p = 0.301) among the vehicle-traumatic AKI-RRT patients relative to non-traumatic AKI-RRT were similar. However, vehicle-traumatic AKI-RRT patients had a lower risk of long-term mortality than non-traumatic group (adjust HR: 0.473, p < 0.001) ( Table 2). In addition, vehicle-traumatic AKI-RRT patients was associated with lower ESKD rates (HR, 0.552; 95% CI, 0.325-0.937; p = 0.028) than non-traumatic AKI-RRT patients in multivariable competing risk models which set death as a competing risk. Furthermore, this lower mortality rate after vehicletraumatic AKI-RRT persisted after adjusting for age, gender and Charlson score Cox proportional analysis (p < 0.001) (Fig. 3).

Annual number of trauma cases
After stratified vehicle-traumatic AKI-RRT patients by injured body part (Supplemental Table 2), the risk of developing ESKD and 30-day mortality after discharge among the vehicle-traumatic AKI-RRT patients relative to non-traumatic AKI-RRT were similar except for soft tissue injury from trauma which had lower incidence of 30-day mortality (HR, 0.714, p = 0.003). Bones (adjust HR: 0.580, p < 0.001) and abdomen injury (adjust HR: 0.607, p < 0.001) from trauma had lower risk of long-term mortality than non-traumatic group.

Discussion
Our study is, to our knowledge, the first to describe the short-term and long-term outcomes of vehicle-traumatic AKI-RRT patients from a large national database. Our study demonstrated that vehicle-traumatic AKI-RRT patients had lower long-term mortality than non-traumatic AKI-RRT patients. However, incident ESKD and short-term mortality were similar between vehicle-traumatic and non-traumatic AKI-RRT patients.

The trend of vehicle-traumatic AKI-RRT .
In an observational study in the United States, the incidence of AKI-RRT increased at an average of 10% per year from 2000 to 2009 9 . In the NHIRD, we found that the proportions of patients with AKI-RRT among vehicle-traumatic patients also increased throughout the years. The older the patients, the higher annual number of trauma cases was found; however, the oldest vehicle-traumatic group had the lowest incidence of AKI-RRT (Fig. 2). These results may indicate that young people had higher severity of trauma and the consequent risk of AKI-RRT. There seemed to be no difference in body distribution of injury on the incident vehicle-traumatic AKI-RRT except for the lowest incidence of soft tissue injury related AKI-RRT.

ESKD and long-term mortality in patients with vehicle-traumatic AKI-RRT . AKI in traumatic
patients is a common complication and associated with significantly higher 30-day and 1-year mortality than traumatic patients without AKI 12 . Although trauma-related risk factors for kidney injury (e.g., direct lesions to the kidneys, shock, ischemia-reperfusion, rhabdomyolysis, exposure to nephrotoxic substances, abdominal compartment syndrome, hemorrhagic shock and sepsis), The rate of long term incident ESKD in patients after AKI-RRT in our study did not significantly differ vehicle-trauma and non-trauma populations. However, compared to non-traumatic AKI-RRT patients, traumatic patients had lower risk of long-term mortality after survival in our cohort. The above results may indicate that the etiology of AKI-RRT may be associated with longterm mortality. On the other hand, the risk of ESKD existed if renal damage happened regardless of trauma or non-trauma. If we consider death as a competing risk, vehicle-traumatic AKI-RRT patients were associated with lower ESKD rates than non-traumatic AKI-RRT patients, which may indicate that vehicle-traumatic AKI-RRT patients had better renal recovery potential. In addition, the vehicle-traumatic AKI-RRT patients with soft tissue injuries may have lower severity trauma which may contribute to better 30-day mortality.
The observed decreased long-term mortality and long-term ESKD in vehicle-traumatic AKI-RRT patients after 30 days can be related to the overall better health status of these patients before the traumatic injury and that perhaps a greater burden of co-morbidity and complications in the more non-traumatic AKI-RRT population with greater baseline debility. This cannot be fully addressed with propensity score-matching involving crosssectional comorbidity data at the time of or before index admission.
One of the few other studies examining long term-outcomes after traumatic AKI examined 40 soldiers who survived traumatic AKI-RRT in that study there was a lower incidence of ESKD (2.5% vs. 10.6%) and a lower long-term mortality (2.5% vs. 25%) rate over a median follow-up of 2.7 years than in our cohort 14 . These differences might be attributed to low incidence rates, younger age (26 ± 6 years), and fewer baseline comorbidities in the military study 14 .
The management of vehicle-traumatic AKI. Management of AKI should be considered in the context of the underlying etiology; for instance, sepsis-associated AKI will focus on source control of infection hemodynamic support and secondary injury prevention 15 , while nephrotoxic will focus on identification removal of the offending agent 16 . Table 2. Incidence and risks for outcome of interest in AKI patients with renal replacement therapy during hospitalization, between vehicle-traumatic patients and their matches. ESKD, end-stage kidney disease; CI, confidence interval. *The multivariable Cox regression model selected covariates from all variables in Table 1 and cardiogenic shock and exposure to nephrotoxic medications by a stepwise procedure. www.nature.com/scientificreports/ However, the management of traumatic AKI is complicated because it commonly involves multiple risk factors for AKI, including direct traumatic injury to the kidneys, hypovolemic shock, ischemia-reperfusion injury, exposure to endogenous and exogenous nephrotoxic substances, and secondary infection. Accordingly an understanding of the underlying cause of traumatic AKI is required to enable its optimal management. In addition, we have demonstrated that that traumatic AKI-RRT patients had lower long-term rates of death or ESKD in competing risk analsysis compared to non-traumatic AKI-RRT patients; this might indicate that severe traumatic AKI has a generally good prognosis justifying aggressive treatment including the use of RRT where indicated.
Follow-up strategy in patients of AKI-RRT . AKI-RRT has been shown to be associated with an increased incidence of ESKD, mortality 17 , coronary events 13 , stroke 18 , and bone fracture 19 among those who survived to hospital discharge. Although guidelines published by kidney disease: improving global outcomes (KDIGO) Clinical Practice Guideline recommend that survivors of AKI be followed up by a nephrologist within 90 days 20 , reports in the literature suggest only a small proportion of patients after severe AKI are followed-up by a nephrologist after survival discharge [21][22][23] . Given their young age, relative lack or comorbidity, and still significant risk of developing ESKD, AKI-RRT is a neglected population where focused medical follow-up could address the progression of CKD and development of other complications over a long expected lifespan. These results further provide the outcome information for patient care and rational health resource allocation.
Limitation. Our study has certain limitations. Firstly, as all observational studies, we could not exclude the possibility of residual confounding. To decrease the effect of potential confounding factors, we developed a 3:1 propensity score. We were able to construct a comparison cohort with balanced covariates between the two groups. Second, with some administrative databases, detailed in-hospital parameters, and laboratory results were not available in our cohort. Therefore, we could not examine whether classifying patients into different severity scoring systems, such as the Injury Severity Score (ISS) and Acute Physiology and Chronic Health Evaluation score. The severity of trauma and illness has been found to be associated with AKI 8,24 , and thus could be potential predictors of clinical outcome. To attenuate the effect of the limitations, we did adjust by Charlson comorbidity index (pre-admission comorbidities) and in-hospital acute comorbidities, including pulmonary and hemodynamic complications, to evaluate the disease severity of each patient in our database. We further stratify our trauma patients based upon injured body regions as proxies to evaluate the traumatic severity. Third, we did not analyze the effect of duration and modality of RRT which may be the predictors of disease severity and the confounding factors of long-term outcomes in our study. To attenuate the effect of the limitations, we did adjust by Charlson comorbidity index (pre-admission comorbidities) and in-hospital acute comorbidities, including pulmonary and hemodynamic complications, to evaluate the disease severity of each patient in our database.

Conclusion
In our study, we found that the vehicle-traumatic AKI-RRT patients had better long-term survival than nontraumatic counterparts with AKI-RRT. However, there were similar outcomes of ESKD and short-term mortality. Our findings are pertinent to civilian populations with similar demographics and may inform care in real-world practice.

Methods
Study design and setting. National Health Insurance registration database (NHIRD). The Taiwan National Health Insurance (NHI) program is mandatory and universal, offering comprehensive medical care coverage to more than 99% of the country's population of 23 million people. This nationwide compulsory healthcare program that covers outpatient visits, hospital admissions, prescriptions, interventional procedures, disease profiles, and vital statuses. Taiwan's NHI records are regularly inspected, and physicians are subject to statutory regulation 13,19,[25][26][27][28] . The National Health Insurance Research Database (NHIRD) thus represents a comprehensive, high-quality record of healthcare episodes provided to the Taiwanese population.

Selection of participants.
We enrolled patients aged ≥ 18 years who were admitted because of major trauma and developed AKI-RRT during their index admission as recognized by RRT-related procedure codes and survived to hospital discharge. All diagnoses, including trauma-related injuries and baseline comorbidities, were obtained by the codes of International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM). Those that were related to vehicle-related accidents, defined as any of ICD-9-CM E-Code E800 to E844 included in the diagnosis codes, were aggregated to individuals and were enrolled as study subjects 25 . When more than one accident was encountered by a single person, only the first event was counted for analysis. We identified baseline comorbidities from at least three outpatient visits or one inpatient claim within a oneyear antecedent to the index admission with first dialysis. This rule was constructed based on a relatively strict criterion and was well validated with good predictive power 13,19,25,26,29 .
Dialysis patients who underwent renal transplantation, vascular access creation, or peritoneal-dialysis catheter implantation for chronic dialysis and those who had received chronic dialysis prior to the index admissions were excluded. To compare the long-term outcomes of interest, we constructed a comparator group of non-traumatic AKI-RRT patients matched to vehicle-traumatic AKI-RRT patients in a 3:1 ratio based on age, gender, and propensity scores. (Supplemental Table 4).
As all personal information is de-identified in the research database to protect privacy, no informed consent was required, and this study was deemed exempt from a full ethical review by the institutional review board of the National Taiwan University Hospital (201212021RINC). www.nature.com/scientificreports/ Outcomes. Our primary outcome was long-term all-cause mortality after hospital discharge. The secondary outcomes were 30-day all-cause mortality and de novo end-stage kidney disease (ESKD) defined as the requirement of dialysis for at least three months after hospital discharge. We used a selection period of 90 days to define ESKD because all patients receiving dialysis for more than 90 days in Taiwan can apply to the NHI for catastrophic illness registration cards 30 . Each patient was monitored from the date of discharge and was censored at either death, dialysis, or the end of the study (December 31, 2010), whichever occurred first. We also stratified trauma patients by injured body part and age to evaluate the outcomes comparing to their matches in the subgroup analysis.
Research variables. We recorded the disease severity and patient condition during index hospitalization.
Disease severity and patient condition are estimated by the ICU procedure and complications with acute pulmonary [prolonged mechanical ventilation (the use of mechanical ventilation ≥ 21 days), re-intubation, acute respiratory distress syndrome, pleural effusion, and chest tube insertion], cardiovascular [hemopericardium, hypovolemic shock, cardiac arrest, heart block, atrial fibrillation, extracorporeal membrane oxygenation (ECMO) and intra-aortic balloon pumping (IABP)] and infectious (pneumonia, urinary tract infection and severe sepsis) and other (delirium, stroke and gastrointestinal bleeding) disorders 31 .
The Charlson comorbidity index 32 was computed using baseline comorbidities. Additional adjustments in these models included control for direct effects from age, gender, and comorbidities that are listed in Table 1.
Analysis. Baseline characteristics were described as the percentages for categorical variables and median with interquartile range (IQR) for continuous variables. Differences between the trauma and non-trauma related groups were compared by the independent t-test or χ 2 test where appropriate.
To estimate each patient's propensity score for vehicle-traumatic AKI-RRT, we fitted a separate multivariable logistic regression model with the factors predicting vehicle trauma as admitting diagnosis in patients with AKI-RRT during index hospitalization 33 (further seen in Supplemental Table 4). The caliper distance is 0.25 and subjects are matched without replacement in this propensity score matching. The estimated propensity score was also added to adjusting the Cox regression model as a single covariate for controlling selection bias. Multivariable logistic regression models before and after propensity-scored matching were applied to estimate odds ratio (OR) of study outcomes after adjusting all the confounders predicting trauma (Supplemental Table 4).
We evaluated the association of vehicle-traumatic AKI-RRT with the risks of long-term ESKD, 30-day mortality and long-term mortality among our cohort using Cox regression models with time-dependent covariates. To assess multivariate Cox regression models in mortality and ESKD analysis, the relationship between scaled Schoenfeld residuals and time was examined, with no significant evidence for non-proportionality (p value > 0.05), therefore we accepted the assumption in the Cox model of constant hazard ratios over time.
The significance levels for entry and stay were set to 0.15 to be conservative. Then, with the aid of substantive knowledge, the best candidate final logistic model was identified manually by dropping the covariates with p value > 0.05 one at a time until all regression coefficients were significantly different from 0.
All the analyses were conducted with R software, version 2.8.1 (Free Software Foundation, Inc., Boston, MA, USA); competing-risk analysis was performed using Stata/MP version 12 (Stata Corporation). A two-sided p-value < 0.05 was considered to be statistically significant.

Scientific Reports
| (2020) 10:20572 | https://doi.org/10.1038/s41598-020-77556-3 www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.