Ureteral Trauma: Preoperative Studies NeitherPredict Injury Nor Prevent Missed Injuries

Diana Medina, MD,* Robert Lavery, MA,* Steven E Ross, MD, FACS,† and David H Livingston, MD, FACS*

Background: Ureteral injuries are uncommon, and thenecessity, accuracy, and optimal use of perioperativetesting remains unknown. Delays in diagnosis have alsobeen associated with significant morbidity, includingloss of renal function.

Study Design: The records of all patients (n 5 20) ad-mitted with ureteral injuries to two Level I trauma cen-ters during a 5-year period were reviewed. Data collectedincluded patient demographics, mechanism of injury,degree of associated injuries, and presence of gross ormicroscopic hematuria. The use of any pre- or intraop-erative testing was specifically noted. The location of theureteral injury was obtained from the operative notes.The morbidity and mortality associated with ureteralinjuries in the primarily diagnosed and the delayedgroups were assessed. Presenting signs and symptoms,diagnostic testing, and the urologic management of thepatients in the delayed group were reviewed.

Results: All patients were men whose ages ranged from15 to 72 years, with a mean age of 29. The mechanismsof injury were gunshot wounds in 15, stab wounds in 4,and blunt vehicular trauma in 1. Excluding other uro-logic injuries, the incidence of hematuria related to theureteral injury alone was 53%. A total of 10 pre- andintraoperative studies were performed, only 2 demon-strated the ureteral injury. Seventeen patients had theirinjuries diagnosed primarily. In this group, the ureterwas repaired by suturing and stenting in 12, suturingwithout a stent in 1 and ureterocystostomy in 4. De-layed diagnosis of their ureteral injuries occurred inthree patients. All three missed injuries occurred in theupper portion of the left ureter. All ureters were success-fully repaired. There were no mortalities in this group,nor did any patient require a nephrectomy.

Conclusions: Direct visualization of the injury is thebest and most accurate diagnostic modality in ureteral

trauma. These results reinforce that a thorough explora-tion of all retroperitoneal hematomas after penetratingtrauma remain an integral part of the total abdominalexploration for trauma. (J Am Coll Surg 1998;186:641–644. © 1998 by the American College of Sur-geons)

The incidence of traumatic ureteral injuries has var-ied depending on the group studied. Presti and asso-ciates1 reported that ureteral injuries comprised lessthan 1% of all urologic trauma; others, who exam-ined series restricted to abdominal gunshot wounds,found that the ureter was injured in up to 4% ofpatients.2-4 This low rate of incidence combined withthe need for urgent celiotomy in most cases has madethe value of pre- and intraoperative testing difficultto assess in patients with suspected ureteral inju-ries.1,4,5 This study was performed to determine theeffectiveness of preoperative testing in the diagnosisof traumatic ureteral injuries. The presentation,management, and outcome of patients with possibleureteral injuries was investigated. Emphasis wasplaced on identifying factors that may contribute todelayed diagnosis in certain patients.

METHODSAn ICD-9 computerized search of patients with uro-logic injuries admitted to theTrauma Services at Uni-versity Hospital (Newark, NJ) and Cooper Hospital(Camden, NJ), two Level I trauma centers, during a5-year period from 1990 to 1995 was performed.The medical records of patients with ureteral injuriesfrom either blunt or penetrating trauma werereviewed.

Demographic data and mechanism of injurywere collected on all patients. The degree of associ-ated injuries were measured using the Injury SeverityScore (ISS) and Penetrating Abdominal Trauma In-dex (PATI). Presenting urologic signs such as gross ormicroscopic hematuria were specifically noted aswere the results of any perioperative radiologic imag-

Received December 8, 1997; Revised January 20, 1998; Accepted January 29,1998.From the *Department of Surgery, University of Medicine and Dentistry, NewJersey Medical School and New Jersey Trauma Center, University Hospital,Newark; and †Department of Surgery, Cooper Hospital, University MedicalCenter, Camden, NJ.Correspondence address: David H Livingston MD, University Hospital E-245,150 Bergen Street, Newark, NJ 07103.

641© 1998 by the American College of Surgeons ISSN 1072-7515/98/$19.00Published by Elsevier Science Inc. PII S1072-7515(98)00108-2

ing. Intraoperative findings were collected from op-erative notes and included the amount of hemoperi-toneum, the duration of the operative procedure,whether any dye (indigo carmine or methylene blue)or radiographic studies were performed, the locationof ureteral injury, the method of repair, and associ-ated abdominal injuries. Patients who had their ure-teral injuries diagnosed during their initial celiotomywere considered to be diagnosed primarily. Thosepatients whose injuries were missed initially wereconsidered to have a delayed diagnosis. The morbid-ity and mortality associated with ureteral injuries inboth the primarily diagnosed and the delayed groupswere assessed. Presenting signs and symptoms, diag-nostic testing, and the urologic management of thepatients in the delayed group were reviewed.

RESULTSTwenty patients with traumatic ureteral injuries wereidentified during the 5-year period. All patients weremen whose ages ranged from 15 to 72 years, with amean age of 29. The mechanisms of injury were gun-shot wounds in 15, stab wounds in 4, and bluntvehicular trauma in 1. Ureteral injuries comprised5% of all genitourinary trauma seen at UniversityHospital, Newark during the 5-year study period.

Six patients presented with gross hematuria andsix with microscopic hematuria. Two patients withgross hematuria had concomitant kidney lacerations,and one patient with microscopic hematuria had abladder injury. Once these other urologic injurieswere excluded, the incidence of hematuria, presum-ably from the ureteral injury alone, was 53% (9 of17). All 19 patients sustaining penetrating injury un-derwent emergency celiotomy, an average of 45 min-utes after admission to the hospital. The patient withblunt trauma was operated on within 8 hours ofadmission.

A total of 10 diagnostic studies were obtained.Five were obtained preoperatively: intravenous py-elogram (IVP) in two, abdominal computed tomog-raphy (CT) scan in two, and a retrograde pyelogram(RPG) in one. Only two (40%) were diagnostic ofureteral injury (one IVP, one CT). The remainderwere interpreted as being negative for injury.

The remaining 13 patients (62%) had their inju-ries diagnosed by surgical exploration and direct vi-sualization of the injury. Intraoperatively, methyleneblue or indigo carmine dye testing was performed inthree cases and was positive in only one. Two intra-operative IVP studies were obtained and both werenegative. The overall sensitivity of perioperative test-ing was 20%.

Associated injuries were common and are listedin Table 1. Only one patient had an isolated ureteralinjury. The mean ISS was 13 (range, 9–27) and PATIwas 23 (range, 8–65). There were no significant dif-ferences in ISS and PATI between those patients hav-ing their ureteral injury primarily diagnosed andthose whose injuries were missed.

In the 17 patients whose injury was diagnosedprimarily, the ureter was repaired by primary repairover a double-J stent in 12. One patient had primarysuturing without a stent and four patients with lowerureteral injuries underwent ureterocystostomy andpsoas hitch procedures. All ureteral repairs weredrained with either closed suction or Penrose drains(or both) at the discretion of the treating surgeon.

Delayed diagnosis of ureteral injuries occurred inthree patients, one of whom had a negative intraop-erative IVP, and all three presented with fever, leuko-cytosis, and either localized or generalized peritonealsigns. The diagnosis was made using CT, IVP, andRPG in one patient each. All three missed injuriesoccurred in the upper portion of the left ureter, andthe original operative record in two specifically men-tions a large retroperitoneal hematoma in that area.All ureters were able to be successfully treated bydirect repair over a stent in one patient or with acombination of direct repair and percutaneous stent-ing in the other two. There were no deaths in thisgroup, nor did any patient require a nephrectomy.Their length of hospitalization was 20–61 days andwas considerably longer than the primarily diag-nosed group.

The incidence of urologic complications in theprimarily repaired group was 24% (4 of 17). Thecomplications were a urinoma that required drainagein three and a persistent urine leak resulting in uro-

Table 1. Associated Injuries

Location No. of injuries

Small bowel 13Colon 6Stomach 4Iliac artery/vein 4Liver 3Pancreas 2Bladder 2Kidney 2Diaphragm 1Inferior vena cava 1

642 Medina et al Ureteral Trauma J Am Coll Surg

sepsis in one. There was a 29% (5 of 17) mortalityrate in this series of patients. Three patients diedfrom sepsis and multiple organ failure, one from on-going hemorrhagic shock, and one from an unex-pected cardiac arrest. In one patient with severe mul-tiple abdominal injuries, the treatment of the ureteralinjury was prolonged and difficult and the subse-quent complications were thought to contribute tothe patient’s death from sepsis and organ failure; thedeath rate associated with ureteral trauma in this se-ries was 5%.

DISCUSSIONIn the present study, the ureteral injuries were rarelyisolated and were most often caused by penetratinginjuries, similar to other series of ureteral trau-ma.2,3,5,6 In addition, multiple intraabdominal inju-ries occurred in 95% of patients, making emergencyceliotomy the most common and most accurate di-agnostic test. As such, preoperative testing was oftennot used nor appropriate and intraoperative identifi-cation of the injury became paramount. Of 18 pa-tients, 13 (72%) who underwent surgical explorationwithout preoperative testing were diagnosed intraop-eratively. This is comparable to other published ratesof 40%–83%.2,4-6 Our data also confirm the findingsof others that gross or microscopic hematuria is anunreliable diagnostic sign because it was only presentin 45% of patients.1,4,7

McGinty and Mendez6 reported that right-sidedand upper ureteral injuries were more common, al-though others8 have found no difference in the loca-tions of ureteral injuries. In this series there was anequal distribution between right and left and upperand lower injuries. More interesting was that all threemissed injuries occurred in the upper portion of theureter. Although exploration of the midureter is eas-ily accomplished by mobilizing the proximal rightcolon or sigmoid colon, exploration of the upperureter in the face of a large hematoma may be moredifficult or incomplete, especially if there is concernabout a concomitant renal injury. We hypothesizethat this is the cause of the three missed injuries inour study. We advocate exploring the upper ureter bybeginning in the midureter and tracing it cephaladtoward the renal pelvis. We do not believe that isola-tion of the renal vessels before exploration or a jux-tarenal hematoma is useful, and in patients with largehematomas of renal origin we prefer to control therenal pedicle from within Gerota’s fascia as advocatedby Atala and coworkers.9

Even when preoperative diagnostic testing is per-formed, the sensitivity is low, as shown in our studyand that of others.1,7 This is in discordance withRober and colleagues2 and Carlton and associates10

who found that pre- and intraoperative IVP ade-quately diagnosed ureteral injuries.

Similar studies have shown a delay in diagnosisranging from 0% to 57%.1,2,4,7,8 In comparison,15% of the patients in our study had a delay in diag-nosis. Brandes and associates4 reported an 8% delayin diagnosis; however, 83% of the patients sustainedinjury to either the middle or lower ureter. The pa-tients with delayed diagnosis in our study all hadupper ureteral injuries. The wide disparity may sug-gest that upper ureteral injuries are less accessible toinspection or less likely to be explored, especially inthe setting of a retroperitoneal hematoma, and maylead to an increase in missed injuries.

Delayed diagnosis of the ureteral injuries hasbeen reported to result in a poor urologic outcome.Although delayed recognition of the ureteral injuriesreported in this study were associated with an in-creased morbidity, no patient had decreased kidneyfunction or required nephrectomy. It is possible thatadvances in percutaneous and endoscopic urologicand radiologic techniques during the past decadehave allowed better control of infection and urinarysepsis and have led to improved renal salvage.

In summary, ureteral injuries remain uncommonand often occur in the patients with multiple abdom-inal injuries. Because both pre- and intraoperativediagnostic testing is inaccurate, direct visualization ofthe ureter is necessary to make the diagnosis. Al-though a delay in diagnosis in this series did notresult in a loss of renal function or late nephrectomy,it did result in infection, urinary sepsis, and pro-longed hospital stays. These data are significant be-cause nonoperative management of selected bluntand penetrating injuries is becoming more common-place.11 A high index of suspicion for a missed injuryand early additional investigation is needed in pa-tients with unexplained abdominal findings inwhom there was an inability or failure to explore theureter in the presence of a retroperitoneal hematoma.

References1. Presti J, Carroll P, and McAninch J. Ureteral and renal pelvic

injuries from external trauma: diagnosis and management.J Trauma 1989;29:370–374.

2. Rober P, Smith J, and Pierce J. Gunshot injuries of the ureter.J Trauma 1990;30:83–87.

3. Bright T, and Peters P. Ureteral injuries due to external violence.J Trauma 1977;17:616–620.

643Vol. 186, No. 6, June 1998 Medina et al Ureteral Trauma

4. Brandes S, Chelsky M, Buckman R, and Hanno P. Ureteral inju-ries from penetrating trauma. J Trauma 1994;36:766–768.

5. Velmahos G, Degiannis E, Wells M, and Souter I. Penetratingureteral injuries: the impact of associated injuries on management.Am Surg 1996;62:461–467.

6. McGinty D, and Mendez R. Traumatic ureteral injuries with de-layed recognition. Urology 1977;10:115–117.

7. Campbell E, Filderman P, and Jacobs S. Ureteral injury due toblunt and penetrating trauma. Urology 1992;40:216–220.

8. Peterson N, and Pitts J III. Penetrating injuries of the ureter. J Urol1981;126:587–590.

9. Atala A, Miller FB, Richardson JD, et al. Preliminary vascularcontrol for renal trauma. Surg Gynecol Obstet 1991;172:386–390.

10. Carlton C, Scott R, and Guthrie A. The initial management ofureteral injuries: a case report of 78 cases. Urol 1971;105:340–355.

11. Renz B, and Feliciano D. Gunshot wounds to the right thoraco-abdomen: a prospective study of nonoperative management.J Trauma 1994;37:737–744.

644 Medina et al Ureteral Trauma J Am Coll Surg