Failure to Obtain Computed Tomography Imaging in Head Trauma: A Review of Relevant Case Law

Transcription

1 ORIGINAL CONTRIBUTION Failure to Obtain Computed Tomography Imaging in Head Trauma: A Review of Relevant Case Law Rachel A. Lindor, MD, JD, Eric T. Boie, MD, Ronna L. Campbell, MD, PhD, Erik P. Hess, MD, MSc, and Annie T. Sadosty, MD Abstract Objectives: The objectives were to describe lawsuits against providers for failing to order head computed tomography (CT) in cases of head trauma and to determine the potential effects of available clinical decision rules (CDRs) on each lawsuit. Methods: The authors collected jury verdicts, settlements, and court opinions regarding alleged malpractice for failure to order head CT in the setting of head trauma from 1972 through February 2014 from an online legal research tool (WestlawNext). Data were abstracted onto a standardized data form. The performance of five CDRs was evaluated. Results: Sixty relevant cases were identified (52 adult, eight children). Of 48 cases with known outcomes, providers were found negligent in 10 cases (six adult, four pediatric), settled in 11 cases (nine adult, two pediatric), and were found not liable in 27 cases. In all 10 cases in which providers were found negligent, every applicable CDR studied would have indicated the need for head CT. In all eight cases involving children, the applicable CDR would have suggested the need for head CT or observation. Conclusions: A review of legal cases reported in a major online legal research system revealed 60 lawsuits in which providers were sued for failing to order head CTs in cases of head trauma. In all cases in which providers were found negligent, CT imaging or observation would have been indicated by every applicable CDR. ACADEMIC EMERGENCY MEDICINE 2015;22: by the Society for Academic Emergency Medicine Head trauma is a common presenting complaint among emergency department (ED) patients. Over one million computed tomography (CT) scans are performed annually in the evaluation of head trauma, with less than 10% demonstrating clinically significant head injury. 1 There is growing interest across medical specialties in reducing unnecessary radiation exposure, and a particular interest within emergency medicine in reducing unnecessary head CTs in the ED, as highlighted by the American College of Emergency Physicians (ACEP) 2013 involvement in the Choosing Wisely campaign. 2 Several clinical decision rules (CDRs) are highly sensitive in identifying patients at increased risk for clinically significant head injury who would benefit from head CT and, similarly, patients in whom head CT can be safely avoided. Despite the longstanding existence of these rules, over 75% of head trauma patients receive head CT in the ED, and up to 35% of these are not indicated by CDRs. 1 Providers have cited fear of lawsuits as one of the reasons for their overreliance on CT imaging. 3 The likelihood that liability concerns are driving some of this practice is supported by data documenting lower head CT rates in states that have undergone some type of tort reform. 4 Despite providers concern for liability, little is currently known about actual lawsuits against providers for failing to order head CTs in cases of head trauma. This study sought to describe lawsuits in which providers have been sued for failing to obtain head CT in cases of head trauma and to determine whether current CDRs, if they had been applied in the From the Department of Emergency Medicine, Mayo Clinic College of Medicine, Rochester, MN. Received April 16, 2015; revision received June 24, 2015; accepted July 28, Presented at the American College of Emergency Physicians Scientific Assembly, October 2014, Chicago, IL. The authors have no relevant financial information or potential conflicts to disclose. Supervising Editor: Timothy Jang, MD. Address for correspondence and reprints: Annie T. Sadosty, MD; sadosty.annie@mayo.edu by the Society for Academic Emergency Medicine ISSN doi: /acem PII ISSN

2 1494 Lindor et al. FAILURE TO OBTAIN CT IMAGING IN HEAD TRAUMA: A REVIEW OF RELEVANT CASE LAW course of clinical care, would have indicated the need for head CT. METHODS Study Design This was a retrospective review of legal case reports included in a subscription-based online legal research tool. The study was deemed exempt from review by the institutional review board. Study Setting and Population An online legal research tool (WestlawNext; was searched for all relevant court opinions, jury verdicts, and settlements. The databases searched included all cases and jury verdicts and settlements. These databases include many smaller databases containing attorney-reported jury verdicts, settlements, judgments, and arbitrations from U.S. state and federal courts as well as summaries of legal cases from all state and federal courts that were reported by judges involved in the cases or by Westlaw attorneys. The reported cases contain information regarding the date, location, and parties of the lawsuits, as well as varying amounts of information about the details giving rise to the lawsuits and the legal outcomes of the suits. It is not possible to determine the number of cases included in each database within Westlaw, but a search of the term malpractice revealed over 10,000 cases (the maximum number of search hits allowed), including 1,803 reported cases from the past 3 years. These databases were searched using a Boolean search with the query terms [ CT scan, CAT scan, head CT, computed tomography, or CT imaging, and epidural, subdural, hemorrhage, hematoma, intraparenchymal, or intracranial, and emergency, ED, ER, or urgent, and malpractice or negligence ]. All cases reported from 1973, the year CT was first used clinically, 5 through January 2014 were included in this study. No medical records were accessed. This search strategy was similar to that used in previous legal case series. 6,7 Study Protocol Per recognized chart review methods, 8 a standardized data collection form was created to record patient characteristics, clinical course, and performance of applicable CDRs. The performance of five CDRs, summarized in Table 1, was evaluated, including ACEP 2009 clinical policy on Neuroimaging and decisionmaking in adult mild traumatic brain injury in the acute setting, 9 the New Orleans Criteria, 10 the National Emergency X-ray Utilization Study (NEXUS) II, 11 the Canadian CT Head Rule (CCHR), 12 and for cases involving minors, the Pediatric Emergency Care Applied Research Network Pediatric Head Injury rules 13 (see Table 1). Two primary abstractors (RAL and ETB) piloted the data collection form by independently abstracting five full cases. Ambiguities in data collection were clarified with the entire investigative team. The two abstractors then independently abstracted the information for all 60 cases, determined which CDRs applied to each case, and determined if the applicable CDRs would or would not have indicated a CT scan. A senior investigator (RLC) adjudicated any disagreements. Outcome Measures Our primary outcome was the number of lawsuits against health care providers for failing to obtain a head CT in cases of head trauma. Our secondary outcome was the result of the lawsuit, categorized as a finding of negligence against the provider, a settlement agreement, a finding of no liability for the provider (either a jury finding or dismissal of the case prior to trial), or an unknown result. Our tertiary outcome was the presence of an indication for head CT by applicable CDRs. This was determined by applying the clinical facts included in the case report to the clinical indicators for head CT included in each of the five CDRs in Table 1. The absence or a clinical sign or symptom in the legal case report was assumed to reflect the actual absence of that sign or symptom and was coded as being not present for purposes of determining whether a CDR indicated the need for CT. Data Analysis We present all continuous data as either means with standard deviation or medians with interquartile range as appropriate for the distribution of the data. Categorical data are presented as percent frequency of occurrence with 95% confidence intervals (CIs). Statistical analyses were performed using version 9.3 of SAS statistical software. RESULTS Case Characteristics Of the 1,066 cases identified in the initial search, 516 were excluded because they were unrelated to head trauma, 296 were not medical malpractice cases, 103 dealt with something other than failure to obtain a head CT, and 52 were duplicate cases. Of the remaining 99 cases, 39 did not have any information about the patients or mechanism of injury to analyze and were excluded, yielding 60 unique cases against providers for failure to obtain a head CT in cases of head trauma. Further details regarding the initial search results and excluded cases is presented in Figure 1. Of the 60 total cases identified, 10 resulted in findings of negligence against the provider, 11 ended in settlement, 27 ended in findings of no liability for the provider, and 12 cases had unknown outcomes. In cases in which providers were found negligent, the median award amount was $2.71 million (n = 8; range = $1.75 million to $10.7 million). In cases that settled, the median settlement amount was $1 million (n = 7; range = $0.09 million to $3.35 million). The median time between the incident leading to the lawsuit and the final case outcome or report was 5.0 years (n = 47; range = 2 to 13 years). Patient Characteristics Of the 60 patients, 52 (87%) were adults (age 18 years), and eight (13%) were children (age < 17 years; Table 2). Isolated subdural hematomas were the most

3 ACADEMIC EMERGENCY MEDICINE December 2015, Vol. 22, No Table 1 Summary of Clinical Decision Rules Evaluated ACEP 2009 Grade A Recommendations Patients: acute blunt head trauma with LOC or amnesia, age 16+, GCS 14+ headache, posttraumatic seizure vomiting, memory deficits age > 60, focal neurologic deficit intoxication, coagulopathy GCS < 15, trauma above clavicles Canadian CT Head Rule (CCHR) Patients: LOC, amnesia, or witnessed disorientation. No anticoagulation. GCS 13+ GCS < 15 at 2 hours, 2+ episodes vomiting retrograde amnesia, age 65+ years suspected open or depressed skull fracture dangerous mechanism (same as ACEP Grade B) sign of basal skull fracture (hemotympanum, raccoon eyes, CSF otorrhea/rhinorrhea, Battle s sign) ACEP 2009 Grade B Recommendations Patients: acute blunt head trauma with LOC or amnesia, age 16+, GCS 14+ focal neurologic deficit, severe headache vomiting, basilar skull fracture 65+ years, dangerous mechanism GCS < 15 (ejection from motor vehicle, coagulopathy pedestrian struck, fall from > 3 ft or five stairs) New Orleans Criteria Patients: injury within 24 hours. LOC or amnesia of event. GCS 15. No anticoagulation evidence of trauma above clavicle headache, vomiting anterograde amnesia, seizure age > 60, intoxication NEXUS II Head CT Rule Patients: acute blunt head trauma altered level of alertness (GCS < 14, delayed response to external stimuli, excessive somnolence, disorientation) abnormal behavior (any inappropriate action displayed by the victim, including excessive agitation, inconsolability, refusal to cooperate, lack of affective response to questions or events, violent activity) evidence of skull fracture (periorbital or periauricular ecchymoses, hemotympanum, CSF leak from nose or ears, palpable step-off, stellate laceration from point source, any injury produced by an object striking a localized region of the skull [such as a baseball bat, pool cue, baseball, etc.]) persistent vomiting 65+ years coagulopathy (any impairment of normal blood clotting such as in hemophilia, secondary to medications (Coumadin, heparin, aspirin, etc.), hepatic insufficiency, and other conditions) neurologic deficit (abnormal GCS, motor deficit, gait abnormality, abnormal cerebellar dysfunction, cranial nerve abnormality) scalp hematoma PECARN (<2 yr) Patients: <2 yr with blunt head trauma. GCS > 13 GCS < 15 palpable skull fracture altered mental status (agitation, somnolence, repetitive questioning, slow response to stimulation) occipital, parietal, or temporal scalp hematoma history of LOC > 4 seconds (CT vs. observation) not acting normally per parent (CT vs. observation) severe mechanism (CT vs. observation) (MVC with ejection, death, or rollover; pedestrian or unhelmeted bicyclist struck by vehicle; falls of > 3 ft; head struck by high-impact object) PECARN (2+ yr) Patients: 2 7 yr with blunt head trauma. GCS > 13. GCS < 15 palpable skull fracture altered mental status (agitation, somnolence, repetitive questioning, slow response to stimulation) history of LOC (CT vs. observation) history of vomiting (CT vs. observation) severe mechanism (CT vs. observation) (MVC with ejection, death of passenger, rollover; pedestrian or unhelmeted bicyclist struck by motor vehicle; falls of > 5 ft; head struck by high-impact object) CSF = cerebrospinal fluid; GCS = Glasgow Coma Scale; LOC = loss of consciousness; MVC = motor vehicle crash. common injury, occurring in 35 (58%) of cases. Just over half the patients involved in the cases died from their injuries (n = 32, 53%) while the remaining patients (n = 28, 47%) suffered permanent neurologic deficits. The amount of time between when it was alleged that the provider should have ordered a head CT and when the patient actually deteriorated or died was < 6 hours in 22 cases (39%), between 6 and 24 hours in 20 cases (36%), and > 24 hours in 14 cases (25%). The cases involving children differed from those involving adults in that epidural hematomas were more common among children (five of the eight, 63% in children; seven of the 52, 13% in adults), and children were less likely to die as a result of their injuries. Table 2 summarizes cases that resulted in findings of negligence, settlement, and cases involving pediatric patients. Inter-rater Agreement The degree of inter-rater agreement was excellent, with an overall concordance rate of 99.2% (95% CI = 98.4% to 99.6%) for the determination of the presence or absence of CDR-specific indications for head CT (1,121 of 1,130 potential indications). Inter-rater concordance was 98.3% (95% CI = 89.9% to 99.9%) with regard to

4 1496 Lindor et al. FAILURE TO OBTAIN CT IMAGING IN HEAD TRAUMA: A REVIEW OF RELEVANT CASE LAW Figure 1. Breakdown of cases identified by initial search. agreement about which specific CDRs applied to each particular case (59 of 60 cases). CDR Performance Of the 60 total cases, every one of the applicable CDRs would have indicated the need for CT in 43 cases (72%; 95% CI = 59% to 81%) if the rule had been applied to the facts as reported, including in 10 of 10 (100%; 95% CI = 72% to 100%) cases in which providers were found negligent, and 10 of 11 (91%; 95% CI = 62% to 98%) cases in which providers settled. The single case that resulted in a settlement in which head CT was not clearly indicated is detailed further in the discussion. The CDRs suggested the need for CT or observation in all eight (100%; 95% CI = 68% to 100%) cases involving children. DISCUSSION Our data show that use of currently available CDRs would have indicated the need for head CT in the majority of reported cases in which providers were sued for failing to order head CT scans, including all reported cases in which providers were ultimately found to be negligent and all cases involving children. In clinical practice, providers often rely on reference tools that summarize CDRs into concise lists of indications for CT. Knowledge of the nuances of the rules, often beyond the detail provided in summarized references, was necessary to identify patients who could have benefited from head CT in our study. For example, one of the NEXUS II criteria for head CT is evidence of skull fracture. The full definition of this criterion, and the definition used to validate the rule, includes any injury produced by an object striking a localized region of the skull (such as a baseball bat, pool cue, baseball, etc.). One of the lawsuits in our series involved a patient who was knocked unconscious after being hit in the back of the head by a softball while running between bases during a softball game. Application of an oversimplified version of the NEXUS II CDR may have led providers to recommend against head CT because the patient had no classic clinical evidence of skull fracture such as raccoon eyes, Battle s sign, or palpable step-off; however, applying the full definition of evidence of skull fracture as defined in NEXUS II, the injury produced by the softball hitting the back of the patient s head provided a clear indication to at least strongly consider head CT, and adherence to the rule may have identified the epidural hematoma that eventually led to the patient s deterioration. Similarly, knowledge of the study population in which the CDR was validated is necessary to apply the CDRs effectively. For example, the CCHR and the New Orleans criteria were validated in patients who were not anticoagulated. The CCHR, the New Orleans criteria, and the ACEP 2009 Grade A recommendations apply only to patients who have lost consciousness, are amnestic, or in the case of the CCHR, had witnessed disorientation. There are no data demonstrating the performance of these CDRs in patients not included in the original derivation and validation studies, including patients who are anticoagulated or who have not lost consciousness. Knowing which rules are applicable to patients is a first and necessary step to using these CDRs appropriately. Although adherence to CDRs could have identified the majority of patients who needed a CT in the lawsuits included in this study, there were 17 cases in which CDRs did not clearly indicate the need for CT. The lack of medical information provided in legal case reports may have contributed to the miss rate. For example, the single case that ended in a settlement in which a head CT was not clearly indicated by the CDRs involved a man in a car accident who suffered multiple facial fractures and experienced left-sided numbness, was scheduled for a CT scan the following day, was discharged home, and decompensated from a skull fracture and associated intracranial bleed shortly thereafter. Given his multiple facial fractures, it is possible that he had some clinical evidence of skull fracture at presentation, such as bruising around the eyes, a palpable stepoff, a cerebrospinal fluid leak, or other sign that would have qualified as an indication for head CT by the CDRs. However, none of these clinical signs or symptoms were recorded in the legal case report so they were assumed to be absent, and this case was one of the 17 counted as missed by CDRs. Although some cases were likely missed by CDRs due to missing data in the legal reports, it is also possible that cases could have been missed even if all of the clinical details were known. In the same case detailed above, the man involved in the car accident reported left-sided numbness in his body, which could have represented a neurologic injury. However, following the NEXUS II criteria, which defines a neurologic deficit as an abnormal Glasgow Coma Scale score, motor deficit, gait abnormality, cerebellar dysfunction, or cranial nerve abnormality, a strict reading of the rule would not have indicated the need for head CT in this patient because isolated sensory symptoms did not meet the definition of a neurologic deficit. This case provides an important reminder that these CDRs are meant to be guidelines to inform providers clinical judgment and not fence posts that prevent the use of common sense in the clinical setting.

5 ACADEMIC EMERGENCY MEDICINE December 2015, Vol. 22, No Table 2 Characteristics of Cases That Resulted in Findings of Negligence or Settlement or Involved Pediatric Patients Year of Incident Year Reported Patient Sex Age (yr) Mechanism Injury Clinical Outcome Time Between ED Evaluation and Deterioration Verdict/ Settlement Amount ($ mil) CT Indicated by All Applicable CDRs? Cases that resulted in a finding of negligence against the provider (6 adult, 4 pediatric) F 67 Fall down stairs Subdural Neuro. injury <3 hours $10.7 Y F 79 Fall in hospital Subdural Death hours $1.8 Y 1991 M 38 Fall from standing Subdural Neuro. injury 6 12 hours $1.9 Y M 74 Fall in hospital Subdural Death 3 6 hours $2.0 Y M 29 Hit by surfboard Epidural Death <3 hours $3.0 Y M 74 Fall from standing Subdural Neuro. injury 6 12 hours $4.4 Y M 9 Fall Epidural Neuro. injury $2.4 Y M 14 Assault Epidural Neuro. injury <3 hours Y F 14 Bicycle accident Epidural Neuro. injury 3 6 hours Y M 0.92 Child abuse Subdural Neuro. injury >1 week $5.0 Y Cases that settled (9 adult, 2 pediatric) 2008 M 74 Fall from standing Subdural Death 3 6 hours Y F 66 MVC Subdural Neuro. injury 6 12 hours $2.0 Y 2013 F 81 Fall from standing Other Death hours $1.0 Y 2004 F 72 Fall from standing Subdural Death $2.8 Y M 48 MVC Other Neuro. injury 6 12 hours $0.3 N M 42 Fall from 2nd Epidural Death 3 6 hours $0.2 Y story F 67 MVC Subdural Neuro. injury >48 hours $3.4 Y F 81 MVC Subdural Death 6 12 hours Y M Fall from standing Subdural Death 6 12 hours Y 2007 M 10 Football injury Epidural Neuro. injury 3 6 hours 0/09 Y 1998 F 11 Fall from vehicle Other Neuro. injury hours Y Cases involving children (ages 0 16 yr) M 9 Fall Epidural Neuro. injury $2.4 Y M 14 Assault Epidural Neuro. injury <3 hours Y F 14 Bicycle accident Epidural Neuro. injury 3 6 hours Y 2007 M 10 Football injury Epidural Neuro. injury 3 6 hours $0.1 Y M 0.92 Child abuse Subdural Neuro. injury >1 week $5.0 Y 1991 M 10 Fall off bike Subdural Death hours $0.0 Y 1998 F 11 Fall from vehicle Other Neuro. injury hours Y M 16 Fall from vehicle Epidural Death 6 12 hours $0.0 Y = value unknown. MVC = motor vehicle crash; neuro. = neurologic LIMITATIONS This study is based on lawsuits that are reported in legal databases. These databases primarily include cases from appeals-level courts and only a small number of cases that are settled out of court or decided in a trial court without appeal. As a result, the number of reported cases represents a small but unknown proportion of the total number of lawsuits levied against providers. Obtaining similar data through an insurance database may allow for better quantification of total number of claims and legal outcomes of additional cases. The data gathered in this study are largely from court opinions that were written by individual judges and do not follow any standards for data inclusion, especially regarding the amount of clinical information included in each case. Both pertinent positive and negative clinical signs and symptoms were often missing from the legal case reports. To not artificially boost the sensitivity of the CDRs evaluated, we assumed that any information not specifically mentioned in a legal case report was absent. This approach likely resulted in an underestimation of the overall sensitivity of CDRs in our study. CONCLUSIONS Providers have reported fear of liability as one of the reasons for ordering head CTs in the evaluation of head trauma, even when not indicated by applicable clinical decision rules. Between 1973 and 2014, 60 lawsuits involving alleged malpractice for failure to order head CT scans for head trauma were identified. Of these, 10 resulted in findings of negligence against the provider, 11 were settled out of court, 27 ended in findings of no liability, and 12 had unknown outcomes. Our retrospective application of five commonly used clinical decision rules demonstrated that use of a clinical decision rule may have indicated the need for head CT in the majority of reported lawsuits against providers and all reported findings of negligence. No cases were identified in which providers were found negligent for failing to order head CT scans when at least one of the applicable

6 1498 Lindor et al. FAILURE TO OBTAIN CT IMAGING IN HEAD TRAUMA: A REVIEW OF RELEVANT CASE LAW clinical decision rules also did not indicate the need for head CT. These data suggest that rather than increasing the risk of liability, adherence to an applicable clinical decision rule by obtaining a CT when indicated may have the potential to decrease the risk of liability for providers. References 1. Melnick ER, Szlezak CM, Bentley SK, Dziura JD, Kotlyar S, Post LA. CT overuse for mild traumatic brain injury. Jt Comm J Qual Patient Saf 2012;38: American College of Emergency Physicians. ACEP Announces List of Tests as Part of Choosing Wisely Campaign. Available at: cal Practice-Management/ACEP-Announces-List-of- Tests-As-Part-of-Choosing-Wisely-Campaign/. Accessed March 14, Rohacek M, Albrecht M, Kleim B, Zimmermann H, Exadaktylos A. Reasons for ordering computed tomography scans of the head in patients with minor brain injury. Injury 2012;43: Smith-Bindman R, McCulloch CE, Ding A, Quale C, Chu PW. Diagnostic imaging rates for head injury in the ED and states medical malpractice tort reforms. Am J Emerg Med 2011;29: Baker HL Jr. Historical vignette: introduction of computed tomography in North America. AJNR Am J Neuroradiol 1993;14: Liang BA, Zivin JA. Empirical characteristics of litigation involving tissue plasminogen activator and ischemic stroke. Ann Emerg Med 2008;52: Lindor RA, Campbell RL, Pines JM, et al. EMTALA and patients with psychiatric emergencies: a review of relevant case law. Ann Emerg Med 2014;64: Gilbert EH, Lowenstein SR, Koziol-McLain J, Barta DC, Steiner J. Chart reviews in emergency medicine research: where are the methods? Ann Emerg Med 1996;27: Jagoda AS, Bazarian JJ, Bruns JJ, et al. Clinical policy: neuroimaging and decisionmaking in adult mild traumatic brain injury in the acute setting. Ann Emerg Med 2008;5: Haydel MJ, Preston CA, Mills TJ, Luber S, Blaudeau E, DeBlieux PM. Indications for computed tomography in patients with minor head injury. N Engl J Med 2000;343: Mower WR, Hoffman JR, Herbert M, et al.; NEXUS II Investigators. Developing a decision instrument to guide computed tomographic imaging of blunt head injury patients. J Trauma 2005;59: Stiell IG, Wells GA, Vandemheen K, et al. The Canadian Head CT Rule for patients with minor head injury. Lancet 2001;357: Kupperman N, Holmes JF, Dayan PS, et al.; Pediatric Emergency Care Applied Research Network. Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet 2009;374: