Penetrating renal injuries: an observational study of non-operative management and the impact of opening Gerota’s fascia

A total of 899 patients were admitted with penetrating abdominal trauma (GSW = 563, Stab = 336). One hundred and fifty patients were diagnosed with a penetrating renal injury (GSW = 54.7%, Stab = 45.3%). Isolated renal trauma was encountered in 50 patients (33.3%). Concurrently injured organs were most commonly the liver (40.7%), diaphragm (25.3%), and colon (24.0%). No resuscitative thoracotomies were performed in either group. Patients sustaining GSW’s were more severely injured, more often presented with peritonitis, had a higher rate of concurrent injury, and more often had unstable vitals (p < 0.05). Demographic and vitals information is given in Table 1.

Table 1 Presenting clinical and demographic data from patients admitted to GSH with penetrating renal injuries from April 30, 2015 to January 30, 2019Initial investigations

Of all patients (n = 899) admitted with penetrating abdominal trauma, 110 (12.2%, 95%CI 10.2–14.6) were reported to have gross hematuria (stab = 26.4% (95%CI 18.4–35.6%), GSW = 73.6% (95%CI 64.4–81.6%)), 42.7% (95%CI 33.3–52.5%)) of which did not have a renal injury. Microhematuria was observed in 332 patients (Stab = 38.6% (95%CI 33.3–44.0%), GSW = 58.4% (95%CI 52.9–63.8%)), 244 (73.5%) of which did not have a renal injury. Overall, any form of hematuria had a 95.3% sensitivity (95% CI 90.5–98.1%) and 60.3% (95%CI 56.7–63.9%) specificity for renal injury, with a negative predictive value of 98.4% (95% CI 96.8–99.2%). Of patients with kidney injuries who required immediate laparotomy, 94.0% (95%CI 90.5–98.1%) had hematuria, 44.7% (95%CI 36.3–53.3%) of which was gross hematuria. The negative predictive value of clear urine (no gross or microscopic hematuria) for ruling out a renal injury in a patient requiring immediate laparotomy for any reason was 98.7% (95%CI 96.3–99.6%).

Of all patients, 496 (55.2%) underwent urgent laparotomy. Immediate laparotomy based on clinical exam alone was performed in 351 patients (70.5%). An immediate laparotomy following admission CT scan was performed in 145 patients (34.0%). Urgent CT scan at the time of admission was deemed appropriate in 426 patients (47.4%).

Of patients with a renal injury (n = 150), 73.3% had an urgent CT scan. Renal injury was diagnosed at the time of emergent laparotomy without CT scan in 40 patients (26.7%). The most common indication for laparotomy in patients with renal injuries was peritonitis (67.8%), followed by radiographic findings at CT scan (18.5%), and hemodynamic instability (12.3%). Of patients with renal injuries taken for emergent laparotomy, 84.6% had sustained gunshot wounds.

AAST grading of renal injuries found on CT scan or intraoperatively are given in Table 2. The most common injuries were grade III (39.3%), and Grade IV (30.0%). Gross hematuria was not associated with higher AAST grade kidney injuries (Grade IV/V) (p = 0.22).

Table 2 American association for the surgery of trauma (AAST) grade of renal injuries in patients with penetrating stab and gunshot wounds (GSW) admitted to GSH during the study periodOperative management

Management decisions and outcomes are shown in Figs. 1, 2 and 3. Overall, 55.3% of all patients with kidney injuries were managed non-operatively. Non-operative management was successful in 91.6% of cases. Patients with gunshot wounds were taken to the OR much more commonly than patients with stab wounds (67.1% vs. 14.7%) (p-value < 0.001). Gunshot wounds were much more likely to cause AAST grade V injuries (p = 0.0001), which in turn were far more likely to be managed operatively compared to all other grades (92.8% vs. 39.7%, p < 0.001). In patients with isolated renal trauma (n = 50), only four (8.0%) required laparotomy, all for open kidney repair, none of which required nephrectomy. In all patients going to the OR who underwent total or partial nephrectomy, 100.0% required additional surgical intervention. Of these, 95.7% required major visceral repair/resection, or damage control surgery. The remaining patient required simple liver packing for an AAST Grade II liver injury and was closed primarily. Patients with isolated renal trauma were caused by stab wounds 90% of the time. Patients with gunshot wounds to the abdomen that cause renal injuries were far more likely to have concurrent intra-abdominal injuries (93.9% vs. 33.8%, p < 0.01). The overall rate of nephrectomy for all patients (OM and NOM) was 13.3% (95%CI 8.3–19.8%). Nephrectomy was far more common in GSW than stab wounds (23.1% vs. 1.5%, p-value < 0.0001). Other interventions included partial nephrectomy (2.0%), primary repair (2.0%), and “other” procedures (exploration only, hemorrhage control with electrocautery, and simple packing) (6.7%). For patients in whom Gerota’s fascia was opened, total nephrectomy was the most common procedure at a rate of 55.6% (95%CI 38.7–72.3%, p < 0.001), partial nephrectomy occurred in 8.3% (95%CI 0.0–17.7%), open repair in 8.3% (95%CI 0.0–17.7%), and “other” procedures (packing, simple exploration, application of energy devices/hemostatics) in 27.8% (95%CI 12.6–42.9%). Patients undergoing total nephrectomy (20) were more likely to have grade IV (8, 40%), or V (12, 60%) injuries (p < 0.001). All patients undergoing partial nephrectomy (3) had AAST grade IV injuries. Primary repair was performed in one patient with a grade IV injury, and two patients with grade III injuries. Hemorrhage control, exploration, or simple packing were performed in three grade IV, three grade III, three grade II, and one grade I injury.

Fig. 1figure 1

Clinical decision pathways and outcomes for all patients with renal injuries. NOM = Non-operative management. Failure of NOM was defined as a need for any abdominal surgical intervention. Successful NOM patients were managed without the need for surgical intervention. All patients in diagnostic laparoscopy group were planned, delayed operations for left sided thoracoabdominal stab wounds. DVT = Deep Vein Thrombosis

Fig. 2figure 2

Clinical decision pathways and outcomes for patients sustaining abdominal stab wounds causing renal injuries. NOM = Non-operative management. Failure of NOM was defined as a need for laparotomy. Successful NOM patients were managed without the need for surgical intervention. Primary surgery is defined as the need for emergent/urgent laparotomy directly from the emergency department prior to admission to a hospital ward/intensive care unit. SSI = Surgical Site Infection

Fig. 3figure 3

Clinical decision pathways and outcomes for patients sustaining abdominal gunshot wounds causing renal injuries. NOM = Non-operative management. Failure of NOM was defined as a need for laparotomy. Successful NOM patients were managed without the need for surgical intervention. Primary surgery is defined as the need for emergent/urgent laparotomy directly from the emergency department prior to admission to a hospital ward/intensive care unit

Non-operative management

Non-operative management was successful in 91.6% of cases. Rates of success of NOM were not different between GSW and stab groups (89.9% vs. 92.8%, p-value = 0.64). Univariate logistic regression for factors predicting failure of non-operative management is given in Table 3. None of the recorded variables were reliable predictors of NOM failure on univariate analysis, and therefore multivariate analysis was not performed. In patients undergoing NOM, five (6.0%) required angioembolization (four kidney, one liver). A total of four (4.8%) patients required percutaneous drainage of collections, all due to abscess or biloma from concurrent liver injuries. A single patient is represented with frank hematuria after discharge with no findings on repeat CT angiogram and spontaneous resolution. No patient required cystoscopy or nephrostomy. Of the seven (8.4%) patients failing NOM, only one required nephrectomy. This patient additionally required damage control for deterioration of a severe liver injury. A single patient underwent an exploration of the kidney only. The remainder failed due to hollow viscous injury (2), hemorrhage from the spleen (1), wound sepsis requiring debridement (1), and concern for ongoing liver hemorrhage (1).

Table 3 Univariate logistic regression for factors predictive of failure in NOM

Comparisons of NOM and OM are given in Table 4. Compared to patients undergoing operative management, patients undergoing NOM had lower mortality rates, hospital length of stay, and complication rates, but were also less severely injured. In patients who were management non-operatively, there were no differences in patients sustaining GSW versus stab wounds with regard to mortality (0.0 vs. 0.0%, p = 1.0), overall complication rate (18.5% vs. 12.5%, p-value = 0.53), Clavien-Dindo III/IV complication rate (11.1% vs. 7.1%, p-value = 0.54), or readmission rate (7.4% vs. 8.9%, p-value = 0.90). The median length of stay was longer in GSW victims managed non-operatively versus stab victims (5.0 vs. 3.0, p = 0.018). Comparisons of GSW and stab outcomes are given in Table 5.

Table 4 Comparisons of outcomes in non-operative management (NOM) and operative management (OM) for patients with penetrating renal traumaTable 5 Subgroup analysis of stab wounds and gunshot wounds managed by non-operative management

留言 (0)

沒有登入
gif