Motor Vehicle Crash Victims Research Articles

Thoracoscopy vs Conventional Autopsy of the Thorax

To analyze the sensitivity and specificity of thoracoscopic autopsy compared with that of conventional postmortem examination. Consecutive sampling, case-series study. A general community referral center and the local Institute of Forensic Medicine. A consecutive sample of 20 fresh cadavers (less than 24 hours old), victims of motor vehicle crashes, gunshot wounds, in-hospital deaths. Introduction of the thoracoscope through the fourth or fifth intercostal space, anterior or midaxillary line. The pleural cavity and mediastinum were examined. Formal postmortem examination was then performed, and the findings of both procedures were recorded, compared, and analyzed. Prior to the beginning of the study, it was hypothesized that thoracoscopic autopsy would reach an overall sensitivity and specificity of at least 85% and enable accurate inspection of the thoracic cavity and mediastinum in search of trauma-related abnormalities in trauma victims or cause of death in in-hospital patients. There was an 85% correlation of both procedures in all cases. The sensitivity of thoracoscopic autopsy was 87% and overall specificity reached 72%. Thoracoscopic autopsy is accurate and easy to perform. It is highly sensitive, especially in deaths due to trauma and following thoracic surgery. Thoracoscopic autopsy is an important tool that allows the diagnosis of major chest abnormalities as well as ruling out postoperative catastrophe, information that is invaluable for the surgeon, as well as all the medical staff and families. Thoracoscopic postmortem examination should be offered to mourning families as a cosmetic and reliable option when informed consent for conventional autopsy is lacking.

Alcohol-Related Injuries Among Adolescents in the Emergency Department

See related editorial, p. 221, "Alcohol and Injury in the Emergency Department: Opportunities for Intervention." Study objective: To determine the frequency of positive alcohol readings in adolescent patients presenting for treatment of injury. Design: Patients aged 10 through 21 years were prospectively enrolled in this descriptive study. Demographic data and information about the injury were collected at enrollment. Blood ethanol concentration was measured with a saliva alcohol assay with a lower detection limit of 10 mg/dL (2 mmol/L). Setting: Enrollment was conducted at four emergency departments, an urban trauma center, an urban children's trauma center, a suburban hospital, and a rural hospital. Enrollment at each facility was conducted during two 24- hour periods for every day of the week (14 days total). Consecutive sampling was used during each enrollment period. Results: We enrolled 295 patients (92% of eligible subjects). Sixty-three percent were male; 74% were white, 19% black, 3% Hispanic, 1% Asian, and 3% from other racial groups. The mean age was 15.6±3.2 years. Fifteen patients (5%) tested positive for ethanol (range, 10 to 120 mg/dL [2 to 24 mmol/L]). Only four of these patients underwent ethanol testing as part of their medical evaluations. Of the 125 subjects aged 17 through 21 years, 14 (11.2%) tested positive for ethanol. Hospital distribution was (number of patients with positive ethanol test results): urban trauma center, 8 of 52; urban children's trauma center, 0 of 91; suburban hospital, 4 of 111; rural hospital, 3 of 41. The highest percentage of positive ethanol test results was found at the urban trauma center, where 15% of total subjects and 22% of subjects aged 17 through 21 tested positive. Injuries related to assaults and motor vehicle crashes were particularly associated with alcohol use. Conclusion: Alcohol is associated with injuries in urban, suburban, and rural settings in the older pediatric population. Alcohol use is underrecognized and should be considered in patients presenting with injuries, especially victims of assaults or motor vehicle crashes. [Meropol SB, Moscati RM, Lillis KA, Ballow S, Janicke DM: Alcohol-related injuries among adolescents in the emergency department. Ann Emerg Med August 1995;26:180-186.]

EMS providers do not accurately note motor-vehicle crash patients with positive serum alcohol concentrations.

Alcohol use is associated with many motor-vehicle crashes and may increase the severity of injury. Because alcohol use also may mask injury, prehospital emergency care providers (PHECPs) may make inaccurate assessments. To assess and triage crash victims accurately, PHECPs must identify recent alcohol use. This study examines the accuracy of PHECPs in identifying motor-vehicle crash victims who had positive serum alcohol concentrations. Retrospective cohort study. Included in the study were motor-vehicle crash victims more than 17 years of age who were conveyed directly to a university medical center emergency department by ground ambulance (n = 372). TIME PERIOD: 15 July 1990 to 15 July 1991. Data sources included ambulance report forms and hospital records. Variables that were abstracted included the Revised Trauma Score (RTS), the PHECPs' impression of alcohol use, and serum alcohol concentrations. Sensitivity, specificity, predictive-value positive, predictive-value negative, and 0.95 confidence intervals (0.95 CI) were calculated for the PHECPs' ability to identify patients with a serum alcohol concentration > 0. The relationship between the RTS and the impressions of alcohol use was analyzed with chi-square testing: a p-value of < 0.05 was considered statistically significant. This study suggests that PHECPs cannot accurately identify crash victims with positive serum alcohol concentrations. Selection bias and retrospective design are significant limitations of this study. Future studies should develop and evaluate methods to improve PHECPs' accuracy in assessing alcohol use in motor-vehicle crash victims.

Changing profiles in spinal cord injuries and risk factors influencing recovery after penetrating injuries.

The changing profiles of spinal cord injuries in South Africa are addressed in this study. A retrospective analysis of 551 patients with spinal cord injury. The cause of injury was motor vehicle crashes in 30%, stab wounds in 26%, gunshot wounds in 35%, and miscellaneous causes 9%. There was a significant shift from stab wounds towards bullet wounds over the last five years. Bullet spinal cord injuries increased from 30 cases in 1988 to 55 cases in 1992, while stab spinal cord injuries decreased from 39 cases in 1988 to 20 cases in 1992. The incidence of spinal cord injuries following a motor vehicle crash showed a declining tendency after a transient increase (28 cases in 1988, 40 in 1990, 31 in 1992). Moreover, the problem of severe septic complications has been investigated and various risk factors for sepsis that might impair the rehabilitation process have been examined. The risk of developing septic complications was higher in gunshot spine injuries (21 cases out of 193) than in knife injuries (5 cases out of 143). The presence of a retained bullet did not seem to increase the chances for sepsis. In seven patients the sepsis was the direct consequence of the retained bullet while in 14 patients sepsis developed with no bullet in situ. Furthermore, the site of the injury (cervical, thoracic, lumbar spine) did not correlate with the abovementioned risks. Gunshots carry a heavier prognosis. Only 32% of our gunshot cases underwent a significant recovery as opposed to 61% of stab cases and 44% of the motor vehicle crash victims.

Geographic Variation in Preventable Deaths from Motor Vehicle Crashes

In Michigan, drivers in rural motor vehicle crashes (MVCs) are twice as likely to die as nonrural drivers: this could be due to variation in the quality of acute trauma care. This study tests the hypothesis that the preventable death rate (PDR) is higher and that anatomic injury severity is lower for rural compared to nonrural MVC fatalities. Retrospective cohort study. Autopsy results from MVC victims of three rural counties and one nonrural county were reviewed. The time period was 1986-1991. Using the Abbreviated Injury Scale, 1985 version (AIS-85), Injury Severity Scores (ISSs) and Anatomical Profile G scores were calculated. Preventability was determined based on ISSs (< 59) and AIS scores in the head region (< 5). Student's t test and the chi-squared test were used for analysis; a p value of < 0.05 was considered statistically significant. 143 rural and 306 nonrural fatalities were analyzed. The rural PDR was 37.1% and nonrural 48.0% (p < 0.05). ISSs and also G scores were significantly different between rural (54.8; -2.1) and nonrural (50.2; -1.2) areas. This study suggests that regional variation in the quality of acute trauma care is not a significant factor in regional variation in MVC mortality.

Wound ballistics: recognizing wound potential. Part 1: Characteristics of missiles and weapons.

In the United States the number of firearm-related deaths in 1989 was almost equal to the number of motor vehicle-related deaths. Trauma nurses could not imagine themselves caring for motor vehicle crash victims without any understanding of speed, vehicle damage, or collision angles. Gunshot wounds are becoming nearly as frequent as motor vehicle crashes, but the mechanism of injury for a gunshot wound is not as widely understood. This article explains the basics of wound ballistics, emergent care of the gunshot wound victim, and medicolegal concerns for the trauma nurse.

From Roadside to Bedside

To determine the current role of rural hospitals and prehospital agencies in the care of motor vehicle crash victims in a remote, rural county prior to the statewide regionalization of trauma care. Specifically, we determined the proportion of crashes that required a response by emergency medical services (EMSs), the timeliness of the response, the proportion of patients treated in local hospitals, and the factors that predicted referral to trauma centers. Population-based retrospective cohort study linking emergency medical services, emergency department, and hospital discharge data to police motor vehicle crash reports and coroner data. Okanogan County, Washington. All motor vehicle occupants, pedestrians or pedalcyclists who were involved in a motor vehicle collision with a response by emergency medical system personnel or the county coroner. None. EMS response times, emergency department and hospital discharge disposition, Injury Severity Scores, hospital length of stay, procedures, deaths. Twelve percent of 669 crashes reported to the police led to the dispatch of EMS. Crashes with EMS involvement were more likely to include occupants without restraints, who were ejected from the vehicle or who had alcohol as a contributing circumstance. The median interval between crash and EMS dispatch was 5 minutes (95th percentile: 40 minutes), the median scene time was 15 minutes (95th percentile: 35 minutes), and the median interval between dispatch and emergency department arrival was 48 minutes (95th percentile: 95 minutes). Among the 210 patients treated by EMSs, 67 (32%) were admitted to local hospitals, and 19 (9%) were referred to a distant trauma center. Of these, 17 were referred directly from the emergency department, one from the scene, and one after local admission. Compared with patients who were admitted locally, referred patients had a significantly higher mean Injury Severity Score (14.4 vs. 5.1), hospital length of stay (9.1 vs. 1.8 days), and rate of operative procedures (37% vs. 9%). Of the 13 crash-related deaths during the year, nine were declared dead at the scene before EMS arrival, three were asystolic at the time of EMS arrival at the scene, and one died at a referral hospital. The linkage of data from police, prehospital agencies, and hospitals can reveal important information about the sequence of health care for trauma patients. The rural hospitals in this county currently play a major role in the stabilization and treatment of motor vehicle crash victims.

Three-point restraints as a risk factor for chest injury in the elderly.

As the number of drivers over the age of 65 years increases, new trauma patterns are emerging. The use of lap-shoulder restraints, in this population, is beginning to play an important role in new findings about their injuries. Restraint use in motor vehicle crash victims, in relation to average AIS score, ISS, and number of hospital and ICU days, was studied at a level I trauma center. The control group consisted of 100 patients between the ages of 35 and 45 years, who were involved in motor vehicle impacts (MVIs) from January 1990 through January 1992. The study group consisted of 120 patients over the age of 65 years who were involved in MVIs from January 1987 through January 1992. In the control group, restraint use was associated with a lower average AIS score--3.39 versus 7.35 for the unrestrained victims. In addition, the restrained victims spent fewer days in the ICU--0.08 versus 0.30 days for the unrestrained victims; and fewer hospital days--1.08 versus 2.0 days for the unrestrained victims. The general pattern of lower average AIS scores in restrained victims was also seen in the study group. However, in the study group, lap-shoulder belt-restrained patients had higher average AIS scores, in the chest region, than unrestrained patients--0.92 (SE 1.60) versus 0.71 (SE 1.47), respectively. These findings indicate that physicians must be aware of restraint patterns of chest injury in elderly patients. Aggressive triage criteria and vigorous pulmonary toilet, especially in patients with rib fractures, sternal fractures, and pulmonary contusions, should become integral parts of trauma resuscitation protocols in elderly patients.

Blunt diaphragmatic and thoracic aortic rupture: An emerging injury complex

Although both blunt diaphragmatic rupture (BDR) and thoracic aortic rupture (TAR) have been extensively discussed, the association of both injuries has been infrequently mentioned. The purpose of this study was to examine the current prevalence and clinical characteris tics of combined BDR and TAR at an adult regional trauma unit. Among 3,886 trauma victims, 69 (1.8%) had a BDR and 44 (1.1%), a TAR. Seven patients (10% of all patients with a BDR) had both injuries. All 7 were victims of motor vehicle crashes and had a mean Injury Severity Score of 35. All TARs were just distal to the origin of the left subclavian artery. Five patients under went repair of both injuries and survived, 1 patient had only the BDR repaired and survived, and 1 died during emergency thoracotomy, for a survival rate of 86%. Five patients had laparotomy and repair of the BDR in the presence of an unrepaired TAR. The TARs were repaired by the clamp-and-sew technique, three of them with primary repair and two with interposition tube grafts. Concomitant BDR and TAR appears to be an emerging injury complex with both diagnostic and therapeutic challenges. The presence of BDR demands a rigorous search for associated TAR.

Incidence and cost of alcohol-involved crashes in the United States

The incidence of alcohol-involved highway crashes (those in which a driver or nonoccupant had been drinking) was estimated from federal data bases. The estimates were adjusted for police underreporting of alcohol involvement. In 1990, 22% of motor vehicle crash victims—1.2 million—were injured in crashes involving alcohol. Over 22,000 of these victims were killed. The comprehensive cost of alcohol-involved crashes was $148 billion in 1990, including $46 billion in monetary costs and $102 billion in lost quality of life. This represents $1.09 per drink of alcohol consumed. Crashes where blood alcohol concentration (BAC) exceeded .10% accounted for 32% of comprehensive crash costs, and crashes with lower positive BAC accounted for another 8%. Excluding drunk drivers and drunk nonoccupants, alcohol-involved crashes caused 8,500 deaths and left 21,000 people permanently disabled and another 605,000 less seriously injured. Averaged across all drinks, other people collectively pay $0.63 in crash costs every time someone takes a drink. A combination of increased public awareness and strong legal sanctions has been effective in reducing the incidence of alcohol-involved driving. The proportion of injuries in crashes that police reported were alcoholinvolved dropped by 37% between 1982–1984 and 1990

Epidemiology of Traumatic Spinal Cord Injury and Acute Hospitalization and Rehabilitation Charges for Spinal Cord Injuries in Oklahoma, 1988–1990

Few injuries result in more profound and long-term disability than traumatic spinal cord injury. This study describes the demographic and epidemiologic characteristics of traumatic spinal cord injury among Oklahoma residents reported to the statewide, population-based surveillance system in 1988-1990; initial acute hospital and rehabilitation charges for 1989 are also included. There was a reported incidence rate of 40 per million population. Based on a reporting sensitivity of 77%, the "true" incidence of spinal cord injury was estimated to be 51 per million population. Motor vehicle crashes accounted for 48% of injuries. Males aged 15-29 years and blacks were at highest risk of injury. Among blacks, the injury rate due to violence was seven times that for whites or Native Americans. Alcohol/drug use was a contributing factor in 39% of injuries and was highest among males aged 20-29 years (58%), Native Americans (57%), and victims of motor vehicle crashes (48%) or violence (51%). The combined initial charges for persons receiving both acute and rehabilitative care ranged from $9,790 to $666,510, with a median of $53,410 per patient; for complete quadriplegia, the combined median charge was $88,585. Despite its low incidence, hospitalization and rehabilitation charges for spinal cord injury in 1989 caused an economic burden of an estimated $8.4 million. While the charges presented were only a small portion of the total costs of spinal cord injury, they further substantiate the need for prevention efforts targeting these debilitating, often permanent injuries. These efforts should target young males and blacks, and should focus on preventing injuries associated with motor vehicle crashes, violence, and alcohol/drug use.

Psychosocial characteristics and follow-up of drinking and non-drinking drivers in motor vehicle crashes.

Eight hundred fifty-four consecutive motor vehicle crash (MVC) victims admitted from August 1, 1986, through August 31, 1989, were prospectively assessed including measurement of blood alcohol concentration (BAC). One hundred six in-hospital interviews were conducted on competent consenting drivers > or = 18 years old; 22.9% (n = 22) of those who were BAC tested (n = 96) were positive for alcohol on admission. The blood alcohol concentration positive [BAC(+)] and the BAC negative (-) drivers differed significantly on the following variables; driver education [BAC(-) > BAC(+): p < 0.01], license suspension < or = 2 years before admission [BAC(+) > BAC(-): p < 0.01], frequency of self-reported intoxication in month before crash [BAC(+) > BAC(-): p < 0.05], driving within 2 hours of drinking < or = 1 month before admission [BAC(+) > BAC(-): p = 0.01] and self-reported driving with BAC > 17 mmol/L < or = 1 month before admission [BAC(+) > BAC(-): p < 0.01]. Follow-up interviews (n = 106) were conducted 1 year after discharge; drivers originally testing BAC(+) were more likely to drive within 2 hours of drinking (p < 0.05), and were more likely to admit to driving with a BAC > 17 mmol/L (p < 0.01). Original BAC(+) drivers were also more likely to report a subsequent MVC in the year following discharge (not statistically significant). There is a need to develop an assessment system to identify high crash-risk drivers and establish rehabilitation programs to reduce crash recidivism.

SAFETY BELT RESTRAINTS AND COMPARTMENT INTRUSIONS IN FRONTAL AND LATERAL MOTOR VEHICLE CRASHES: MECHANISMS OF INJURIES, COMPLICATIONS, AND ACUTE CARE COSTS

A 3-year prospective study examined 76 frontal (F) and 45 lateral (L) motor vehicle crash (MVC) patients with regard to seatbelt restraint use and occupant compartment contact and intrusion injuries. These 121 MVC victims with multiple injuries (39 belted [B] and 82 non-belted [NB]), admitted to a level I trauma center, were studied by accident reconstruction and medical data analysis. They had a MVC mean impact velocity (delta V) of 30 +/- 11 mph and an injury Severity Score of 29 +/- 12. Proper restraint use reduced brain injury in F MVCs (30% FB vs. 47% FNB) but had no effect in L MVCs (63% LB vs. 30% FB [p < 0.06]). Belt use did not protect against lung, liver, spleen, pelvis, or lower extremity (LE) injury. These appeared to be more a function of crash direction, with LE injuries higher in F crashes (p < 0.04) and pelvis injuries (p < 0.001) higher in L crashes. In FB crashes, however, properly used safety restraints were the primary cause of bowel or colon injuries (p < 0.006). Belts did not prevent thoracic or abdominal solid organ injuries in L crashes. Contact-intrusions (CI) of the car occupant compartment in F crashes were the main cause of brain (A-pillar), lung and liver (steering wheel and instrument panel), and LE (toepan) injuries; but in L crashes, side-door CI caused lung, aorta, liver, and pelvic injuries. In contrast, contact-only (CO) injuries of the steering assembly were mainly responsible for injuries to the lung, heart, and liver in F crashes, and side-door CO for lung, liver, and spleen injuries in L crashes. Deaths and complications after injury were higher among F MVC occupants, or when delta V was > or = 30 mph. Hospital and professional costs reflect the complex care needed for victims of multiple injuries: FB, $99,000; FNB, $95,000; LB, $75,000; LNB, $79,000; total, $10.7 million. Present vehicle safety standards are not adequate, and structural design changes are needed to improve restraints and protect occupants from intrusion-related injuries.

Morbidity among pediatric motor vehicle crash victims: the effectiveness of seat belts.

It is well established that seat belts reduce mortality and morbidity among children. Data are presented for 413 children injured severely enough in motor vehicle crashes to require hospitalization. Of the unrestrained children, 4.5% died, compared with 2.4% of the belted children. Unrestrained children had a higher proportion of injuries in four of five anatomical regions, were more severely injured, stayed longer in the hospital, and were 15% more likely than belted children to be discharged with impairments.

Bilateral Tibial Fractures in Properly Restrained Toddlers Involved in Motor Vehicle Collisions

It is well documented that the use of child restraints, specifically child safety seats, can reduce morbidity and mortality in young victims of motor vehicle crashes. Literature review reveals few extremity fractures in properly restrained toddlers. Of 10 restrained infants involved in car crashes admitted to our hospital, 4 had leg fractures. We report 2 cases of bilateral tibial fractures in restrained toddlers in which the usual clinical signs of fracture were minimal or lacking. A high index of suspicion must be maintained in evaluating even properly restrained toddlers in motor vehicle crashes.

Diagnosis of cervical spine injury in motor vehicle crash victims: How many x-rays are enough?: MacDonald RL, Schwartz ML, Mirich D, et al J Trauma 30: 392–397 Apr 1990

Diagnosis of cervical spine injury in motor vehicle crash victims: How many x-rays are enough?: MacDonald RL, Schwartz ML, Mirich D, et al J Trauma 30: 392–397 Apr 1990

Blood alcohol testing of motor vehicle crash admissions at a Regional Trauma Unit

Motor vehicle crashes remain a leading cause of death and injury in the industrialized world. Alcohol consumption is implicated as a major factor in fatal motor vehicle crashes (MVCs), but only poor estimates of blood alcohol concentrations among nonfatally injured crash victims are available. A 3-year study was undertaken at a Regional Trauma Unit to determine the demographics, injury severity, and alcohol positivity of motor vehicle crash victims. Between August 1, 1986 and July 31, 1989, 825 motor vehicle crash victims were available for study; 368 drivers were admitted to the unit within a period of 4 hours. Of 715 patients tested for alcohol, 31.0% were positive. A total of 333 drivers were tested for blood alcohol; 128 (38.4%) were positive. The mean blood alcohol concentration (BAC) at admission for the drivers was 145.6 mg/100 ml; the estimated mean BAC at crash was 180.9 mg/100 ml. The mean age of BAC positive drivers was 31.4 years, compared to a mean age in the BAC negative drivers of 35.2 years (p less than 0.02). Male patients represented 76.6% of the drivers, yet represented 83.6% of the BAC positive drivers (p less than 0.05). There was a marked seasonal variation in BAC positivity, with 46.1% of drivers positive during the summer months. Alcohol appears to be a significant factor in nonfatal MVCs.

Diagnosis of cervical spine injury in motor vehicle crash victims

As delay in diagnosing unstable cervical spine injuries unnecessarily exposes patients to risk of neurologic injury, it is often recommended that complex radiologic investigations be performed on alert patients with neck pain, tenderness, or neurologic deficit despite normal plain radiographs. The optimal investigation of patients unable to reliably provide such information is less clear. How many X-rays are enough to clear the cervical spine? In order to answer this question, a retrospective review of 775 motor vehicle crash (MVC) victims was conducted. Ninety-two (12%) sustained cervical spine injury. Sixteen of these injuries were missed initially and, in a further 18 cases, the lateral cervical spine X-ray was wrongly interpreted as positive. Fifty-five per cent of patients with cervical injury had a Glasgow Coma Score (GCS) of less than 15 on admission. Lateral radiographic visualization of the complete cervical spine (including a swimmer's view as required) had a sensitivity of 83% and a specificity of 97%. The addition of open mouth (OM) and anteroposterior (AP) views detected all patients with unstable fractures except one man with a head injury who was unable to provide clinical clues to the diagnosis, but who suffered no additional harm as a result. A single lateral X-ray of the cervical spine is inadequate to exclude cervical spine injury in severely traumatized patients and the addition of OM and AP views still failed to identify unstable fractures in one of 385 patients in this series of MVC victims with GCS less than 15.(ABSTRACT TRUNCATED AT 250 WORDS)

Diagnosis of Cervical Spine Injury in Motor Vehicle Crash Victims

As delay in diagnosing unstable cervical spine injuries unnecessarily exposes patients to risk of neurologic injury, it is often recommended that complex radiologic investigations be performed on alert patients with neck pain, tenderness, or neurologic deficit despite normal plain radiographs. The optimal investigation of patients unable to reliably provide such information is less clear. How many X-rays are enough to clear the cervical spine? In order to answer this question, a retrospective review of 775 motor vehicle crash (MVC) victims was conducted. Ninety-two (12%) sustained cervical spine injury. Sixteen of these injuries were missed initially and, in a further 18 cases, the lateral cervical spine X-ray was wrongly interpreted as positive. Fifty-five per cent of patients with cervical injury had a Glasgow Coma Score (GCS) of less than 15 on admission. Lateral radiographic visualization of the complete cervical spine (including a swimmer's view as required) had a sensitivity of 83% and a specificity of 97%. The addition of open mouth (OM) and anteroposterior (AP) views detected all patients with unstable fractures except one man with a head injury who was unable to provide clinical clues to the diagnosis, but who suffered no additional harm as a result. A single lateral X-ray of the cervical spine is inadequate to exclude cervical spine injury in severely traumatized patients and the addition of OM and AP views still failed to identify unstable fractures in one of 385 patients in this series of MVC victims with GCS less than 15. The cervical spine may be cleared in MVC patients by excellent, complete lateral visualization of the cervical spine plus OM and AP views at the risk of missing significant fractures in fewer than 1% of patients.

Diagnosis of Cervical Spine Injury in Motor Vehicle Crash Victims

As delay in diagnosing unstable cervical spine injuries unnecessarily exposes patients to risk of neurologic injury, it is often recommended that complex radiologic investigations be performed on alert patients with neck pain, tenderness, or neurologic deficit despite normal plain radiographs. The optimal investigation of patients unable to reliably provide such information is less clear. How many X-rays are enough to clear the cervical spine? In order to answer this question, a retrospective review of 775 motor vehicle crash (MVC) victims was conducted. Ninety-two (12%) sustained cervical spine injury. Sixteen of these injuries were missed initially and, in a further 18 cases, the lateral cervical spine X-ray was wrongly interpreted as positive. Fifty-five per cent of patients with cervical injury had a Glasgow Coma Score (GCS) of less than 15 on admission. Lateral radiographic visualization of the complete cervical spine (including a swimmer's view as required) had a sensitivity of 83% and a specificity of 97%. The addition of open mouth (OM) and anteroposterior (AP) views detected all patients with unstable fractures except one man with a head injury who was unable to provide clinical clues to the diagnosis, but who suffered no additional harm as a result. A single lateral X-ray of the cervical spine is inadequate to exclude cervical spine injury in severely traumatized patients and the addition of OM and AP views still failed to identify unstable fractures in one of 385 patients in this series of MVC victims with GCS less than 15.(ABSTRACT TRUNCATED AT 250 WORDS)