Incidence of, Risk Factors for, and Mortality Associated With Severe Acute Kidney Injury After Gunshot Wound

IMPORTANCE Acute kidney injury increases the risk of mortality in hospitalized patients. However, incidence of severe acute kidney injury (SAKI) and its association with mortality in civilians with gunshot wounds (GSWs) is not known. OBJECTIVE To determine the incidence of and risk factors associated with SAKI and acute kidney injury requiring dialysis (AKI-D) after GSWs and the association of SAKI and AKI-D with mortality among civilians in the United States. DESIGN, SETTING, AND PARTICIPANTS This retrospective cross-sectional study included civilians withGSWreportedtotheNationalTraumaDataBankbetweenJuly1,2010,andJune30,2015.Torso GSWs were included in study; GSWs to the head were excluded. The data were analyzed between September and November 2018. EXPOSURE Civilians with GSW. Incidence of SAKI and AKI-D; association of SAKI and AKI-D with mortality. factors between patients with and without SAKI, a 1:1 propensity score matching analysis was used to evaluate the association of SAKI with all-cause mortality in patientswithtorsoGSWs. 11 Pairsofthesegroupsofpatientswiththegreedyneighborapproachwere constructed. A caliper setting of 0.01 confirm that the present study was representative of population of civilians with GSWs in the United States.


Introduction
Firearm-related injury caused 38 658 deaths in United States in 2016. 1 Civilian gunshot wounds (GSWs) are the cause of approximately 30 000 hospitalizations and 2500 in-hospital deaths every year in the United States. 2 Survivors of GSWs experience long-term morbidity from injury to various organ systems, such as spinal cord injury, 3 colonic injury, 4 and amputation from vascular injury. 5 Like other organ systems, the urinary tract can be affected by GSW, resulting in acute kidney injury (AKI) from direct injury to the urinary tract or acute tubular necrosis from hypotension, rhabdomyolysis, or sepsis. Acute kidney injury is known to increase the morbidity and mortality of critically ill patients in the intensive care unit setting. 6 Previous research 7 has reported 22% all-cause mortality in US military service members after traumatic AKI that required dialysis. However, the incidence of severe AKI (SAKI) and AKI requiring dialysis (AKI-D) after GSW and the association of SAKI and AKI-D with mortality after GSW in the US civilian population have not been studied. In this article, we describe the incidence of SAKI and AKI-D after GSW in civilians in the United States, the incidence and types of urinary tract injury after GSW, the factors associated with SAKI and AKI-D after GSW, and the association of SAKI and AKI-D after GSW with mortality.

Methods
This is a retrospective cross-sectional study of civilians residing in the United States with GSW reported to the National Trauma Data Bank 8 (NTDB), a registry of patients, including those with GSWs, admitted to trauma centers in United States. Thus, GSW injuries that resulted in death before hospital arrival were excluded from NTDB, and only patients with GSWs who were transported to trauma centers alive were included in this study. Clinical information is stored in NTDB in deidentified form and available for quality improvement and research. All patients in NTDB with GSW to the torso between July 1, 2010, and June 30, 2015, were included in the study. These dates were chosen to keep the study contemporary and avoid the transition from International Classification of Diseases, Ninth Revision (ICD-9) to ICD-10. The data were analyzed between September and November 2018.
The study was approved by the Cook County Health institutional review board with exempt status.
Because this study analyzed deidentified information from the NTDB, individual patient consent was not required. The authors followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for observational studies.
The primary aim of the study was to define the incidence of SAKI and AKI-D, which were identified by the trauma centers reporting data to NTDB. We used the AKI code in NTDB to identify patients with AKI. In NTDB, AKI was defined as an increase in serum creatinine levels of at least 300% of baseline, an increase in serum creatinine levels to more than 4 mg/dL (to convert to micromoles per liter, multiply by 88.4), a decrease in estimated globular filtration rate to less than 35 mL/min per 1.73 m 2 body surface area, reduction of urine output to less than 0.3 mL/kg/h for more than 24 hours, anuria for more than 12 hours, or the need for renal replacement therapy. The definition of AKI in NTDB corresponds to stage 3 AKI according to the Acute Kidney Injury Network (AKIN) classification. 9 Thus, patients with AKI in NTDB who were included in this study experienced SAKI.
We identified AKI-D with ICD-9 codes 39.95, V45.1, V56.0, and V56.1. 10 We excluded patients with a diagnosis of end-stage renal disease who were receiving dialysis at admission (ie, ICD-9 code 585.6 or codes 39.95, 54.98, V45.1, V56.0, and V56.1 without a diagnosis of AKI). We also excluded patients with missing information and patients with preexisting chronic kidney disease (which was recorded in the NTDB file RDS_COMORBID as COMORKEY = 9). We used the following ICD-9 codes to identify injury to urinary tract: 866.0 for injury to the kidney, 867.2 and 877.3 for injury to the ureter, and 867.0 and 877.1 for injury to the bladder and urethra. Demographic information (ie, age, sex, race/ethnicity) and clinical information (systolic blood pressure, heart rate, respiratory rate, Glasgow Coma Scale score, sepsis, and hollow viscus injury) and injury severity score were obtained from the NTDB. Information on all-cause in-hospital mortality, hospital length of stay, intensive care unit length of stay, and length of ventilation was collected from NTDB. All data reported in the current study represent data from the index hospitalization after GSW.

Statistical Analysis
For descriptive variables, we described the mean and SD for continuous variables and number and percentage for categorical variables. Incidence of SAKI and AKI-D was calculated by dividing the number of patients with SAKI and AKI-D by the total number of patients with GSWs. We reported the number and percentage of patients with injury to the urinary tract. For comparison of risk factors for SAKI, AKI-D, and mortality, we used the t test for continuous variables and the χ 2 test for categorical variables. A value of P < .05 was considered statistically significant, and all tests were 2-tailed. The significant variables in univariate analysis were included in a subsequent multivariate logistic regression model using the enter method. The collinearity analysis was also performed, and multicollinearity was defined as a variance inflation factor value greater than 10. With multivariate logistic regression analysis, independent factors associated with SAKI and AKI-D were identified. The association of SAKI and AKI-D with mortality was also assessed with multivariate logistic regression.
Variables that were associated with SAKI were also likely to be associated with increased mortality. To adjust for clinically confounding factors between patients with and without SAKI, a 1:1 propensity score matching analysis was used to evaluate the association of SAKI with all-cause mortality in patients with torso GSWs. 11 Pairs of these groups of patients with the greedy neighbor approach were constructed. A caliper setting of 0.01 was used. Standardized differences were used to confirm a balanced matching result. The matching result was considered balanced when the SD was less than 0.1 12 All original files of NTDB with required data were merged and analyzed with R statistical software version 3.3.1 (R Project for Statistical Computing). Excel version 16.13.1 (Microsoft Corp) was used for data entry and to draw associated figures.

Results
Between July 1, 2010, and June 30, 2015, there were 68 251 patients with torso GSWs in NTDB. Of these, 74 (0.1%) were excluded because they were receiving long-term maintenance dialysis and 4118 (6.0%) were excluded because of missing data. Demographic and clinical variables of the study population are presented in Table 1

Discussion
In this retrospective analysis of civilians with GSWs in United States, the incidence of SAKI was 2.3%, and the incidence of AKI-D was 0.9%. To our knowledge, this is the first study to report the incidence of SAKI and AKI-D after a GSW injury in civilians in the United States. Prior studies have reported an AKI incidence of 1.3% in civilian patients with renal trauma, but they included patients with all types of trauma. 13 Compared with an SAKI incidence of 2.3% in GSWs in civilians in the United States, In the current study, 15.5% of patients had penetrating injury to the urinary tract, and kidney was the most commonly injured organ (63.1%), followed by bladder and urethra (28.1%). Overall, 11.0% had ureteral injury. In a study of battlefield wounds during the war in Croatia, the rate of penetrating urinary tract injury was only 2.6%, with kidney injury predominating (45.1%), followed by bladder (16.5%) and ureter (7.8%). 16 The rate of penetrating injury to the kidney (9.9%) in the current study was also higher than the 5.7% previously reported rate of kidney injury in penetrating trauma. 17 Acute kidney injury following trauma is thought to be because of acute tubular necrosis from ischemia, sepsis, or rhabdomyolysis. Sepsis, hollow viscus injury, Glasgow Coma Scale score, and injury severity score were associated with SAKI. Patients in the current study who developed AKI-D had lower systolic blood pressure, higher incidence of sepsis, and a higher injury severity scale score than patients with AKI not requiring dialysis. Similar findings have been reported in previous studies of AKI in patients who underwent trauma. 15,18 This suggests that acute tubular necrosis owing to ischemia or sepsis was likely the main etiology of AKI in this population.
Acute kidney injury was associated with a significant increase in mortality in the current study.
Mortality in military personnel with AKI-D was reported to be 22%, 7 compared with 28.4% among civilians with GSW and AKI-D in this study. Patients with SAKI were twice as likely to die as patients who did not have SAKI, and patients with AKI-D were 3.2 times more likely to die than patients with no SAKI or AKI-D. This suggests that the severity of AKI is associated with increased mortality in patients with GSW. Acute kidney injury is known to increase the risk of mortality among critically ill patients in the intensive care unit. 6 Chertow et al 19 reported that an increase of creatinine by 0.5 mg/dL or more increased the odds of mortality 6.5 times and resulted in a mean 3.5-day increase in hospital stay. It is now well recognized that even modest increases in creatinine levels increase the risk of mortality and are associated with subsequent development of chronic kidney disease. 20

Limitations
This study has limitations. The NTDB is a convenience sample and thus subject to selection bias, inconsistent measurement of variables, quality of care differences among reporting hospitals, and missing data. However, the outcomes of interest in this study (SAKI, AKI-D, and mortality) were hard clinical end points and less likely to be affected by subjective interpretation. Because this is an observational study, residual confounding is a limitation; however, we have included known variables associated with AKI in this study. Only 6% of the eligible study population was excluded because of missing data, and we feel that the present study was representative of population of civilians with GSWs in the United States.
Acute kidney injury is known to increase future risk of chronic kidney disease. However, the available data did not allow us to estimate risk of chronic kidney disease in patients with GSW who developed AKI, and this needs further research. A total of 588 patients required dialysis in this study; however, we do not have information on the actual indications for dialysis (eg, hyperkalemia, uremia, metabolic acidosis volume overload), dialysis modality (eg, intermitted hemodialysis, continuous renal replacement therapy, slow low-efficiency dialysis), or the association of timing of dialysis (ie, early vs late) with outcomes. A prospective study of patients with AKI after GSW would be able to answer these questions.

Conclusions
In summary, SAKI is a significant problem in civilian patients with GSW and is associated with in-hospital morbidity and mortality. Alignment of surgical guidelines for AKI with AKIN and KDIGO criteria for diagnosis of AKI may result in earlier recognition and management of AKI and possibly lead to better outcomes.