Missed Diagnosis of Spinal Epidural Abscess
A 68-year-old female went to the hospital ED on May 8 complaining of left flank pain following recent removal of a left ureteral stone by lithotripsy. An intravenous pyelogram (IVP) was negative for stones, a blood culture grew E. coli, and a urine culture grew E. coli < 10,000 CFU/ml. She was placed on antibiotic therapy and continued on Coumadin for chronic atrial fibrillation.
On May 10 she was sent back to the hospital by her physician because of worsening left flank pain. She was afebrile, pulse 92, BP 130/70, and respirations 18. A complete blood count (CBC), urinalysis, and blood/urine cultures were ordered. She was admitted by hospitalist A with a diagnosis of probable acute pyelonephritis, placed on IV fluids, and given Dilaudid and Zosyn.
On May 11 the blood culture grew E. coli; the urine culture was negative. The urinalysis report was missing and wasn’t reordered. That evening the patient fell in the bathroom, subsequently stating that her legs “gave way.” The nurse reported this to on-call hospitalist B, who ordered fall precautions but did not examine the patient.
On May 12 at 11:00 AM, hospitalist A saw the patient and learned of the fall. He assumed it was caused by the Dilaudid and reduced the dosage; he did not perform a neurological exam. At 12:30 PM she fell again, and hospitalist A ordered a head CT scan, which was negative; he did not see the patient. She was seen by physical therapy (PT) at 3:30 PM, who charted she was unable to stand and had diminished left lower extremity (LLE) strength (2/5 vs. 4/5 on the right).
At 5:30 PM the nurse charted that the patient was confused, incontinent, and complaining of severe pain in her back and legs. On-call hospitalist C was notified at 7:00 PM about the PT and nursing notes and gave a phone order to add Percocet and Haldol. At 9:45 PM a nurse charted that the patient had bilateral leg numbness, external rotation of the left leg, and “electric shock” pain in the groin. On-call hospitalist C examined the patient and charted that the patient had good strength in both legs with normal sensation.
On May 13 she complained of severe lower back and left thigh shooting pain. At 3:00 PM, PT charted that the patient was unable to complete her exercises because of pain and bilateral leg weakness. Hospitalist A was notified about the PT findings and ordered a lumbar spine x-ray, which was negative.
On May 14 at 12:15 AM the patient became more confused, hallucinated, could no longer lift her feet, complained of shooting pains in both legs, and was unable to plantar flex on the left. She remained incontinent. She was seen by hospitalist D, but her neurological status was not assessed. At 9:00 AM she was seen by hospitalist A, who noted that the international normalized ratio (INR) was 5.1. The Coumadin was discontinued, and she was given vitamin K. He consulted a neurologist by phone, and a stat cervical and thoracic spine MRI was ordered.
The neurologist saw the patient at 8:00 PM and felt she was developing a progressive T4 myelopathy. The stat MRI had not been performed, so he reordered it ASAP. His differential included epidural hematoma due to the episodes of falling with an INR of 5.1 and epidural abscess. On-call hospitalist D called a neurosurgeon, but the neurosurgeon was unwilling to operate because of the INR of 5.1 and the absence of an MRI.
At 10:30 PM the MRI was attempted, but the patient was unable to cooperate and the study was nondiagnostic. Hospitalist D called the neurologist but spoke with the on-call physician, who said a repeat MRI could wait until morning.
On the morning of May 15, an MRI showed discitis with osteomyelitis at T7-8, with an epidural abscess compressing the spinal cord. The patient was taken to surgery for spinal decompression and fusion. After a lengthy rehabilitation, she remained paraplegic with bowel and urinary incontinence.
What Is the Standard of Care?
Both defense and plaintiff experts opined that the patient had an epidural abscess at admission caused by the E. coli bacteremia, which probably originated from the urinary tract. The physicians failed to consider epidural abscess in their differential diagnosis—and to order an MRI of the spine and a neurology consult. The standard of care required the hospitalists to read nurses’ notes and PT assessments and to conduct thorough neurological examinations to evaluate the evolving neurological deficits.
When the patient was admitted on May 10, hospitalist A failed to review the ED record containing the urine culture report < 10,000 CFU/ml and did not perform a neurological examination. On May 11 he failed to reorder the missing urinalysis, which would not have supported the diagnosis of acute pyelonephritis—and when the admission urine culture was negative, he should have considered the possibility of epidural abscess. On May 12 he failed to examine the patient after receiving a call about the second fall and failed to conclude that the patient was deteriorating neurologically.
Hospitalist B failed to review the chart and nursing entries from admission and the ED visit. He should have examined the patient on May 11 when her legs “gave way” and she fell.
Hospitalist C should have examined the patient when first notified on May 12 of her 2/5 LLE weakness and bowel and bladder incontinence. When he examined her later that night, he failed to detect the neurological deficits in her legs.
Hospitalist D’s failure to do a neurological assessment, order an MRI, and obtain a neurology consult after midnight on May 14 delayed the diagnosis by nine hours, thereby decreasing the patient’s chance to regain neurological function.
Should This Claim Be Tried?
While the admission diagnosis of acute pyelonephritis was reasonable, the urine culture was negative, and when the pain became more severe and progressive neurological symptoms developed, spinal epidural abscess should have been considered. An MRI of the spine and a neurology consultation would have led to the diagnosis, and definitive treatment could have been performed prior to developing irreversible neurological deficits. Despite progressive neurological findings, the insureds failed to order an MRI or neurology consult until May 14. Defense expert reviews were nonsupportive for both the standard of care and causation. At the hospitalist group’s request, the claim was resolved.
What Can Be Learned from This Claim?
Spinal epidural abscess (SEA) requires prompt diagnosis and treatment to prevent serious neurological complications. While infrequently encountered in clinical practice, SEA is not infrequently seen in malpractice claims. The alleged negligence is usually a delay in diagnosis resulting in paraplegia with urinary and/or bowel incontinence. The delay typically results from failure to consider SEA in the differential diagnosis in a patient with progressive neurological deficits and severe low back pain. Some claims result from a delay in obtaining an MRI when SEA is considered, either because the hospital doesn’t have an MRI and there is delay in transferring the patient to another facility or because an MRI can’t be done at night or on weekends.
SEA is rare among the many causes of back pain. Fever is an important diagnostic clue because it is not present in most cases of musculoskeletal back pain, such as herniated disc. Fever in a patient with severe, localized back pain, especially if the pain is worsened by percussion, suggests the diagnosis of SEA. Routine laboratory studies are seldom helpful; the leukocyte count may be elevated or normal. Once the diagnosis is considered, spinal imaging is imperative. MRI is preferred because it is positive early in the infection. CT scanning with gadolinium contrast is an acceptable alternative. Plain radiographs of the spine may reveal osteomyelitis or discitis but are rarely diagnostic of SEA.
The classical diagnostic triad consists of fever, spinal pain, and neurological deficits. However, few patients have all three components at presentation. Over time, if untreated, symptoms progress in a typical sequence:
- Back pain, often focal and severe, then
- Root pain, described as “shooting” or “electric shocks” in the distribution of the affected nerve root, then
- Motor weakness, sensory changes, and bladder or bowel dysfunction, and then
- Paralysis, which may quickly become irreversible. Irreversible paraplegia occurs in up to 22 percent of patients, and recovery is unlikely if paralysis is present for more than 24 hours prior to surgery.
SEAs typically begin in the thoracolumbar spine as an infection involving the vertebral disc or the junction between the disc and the vertebral body. The abscess then extends longitudinally in the epidural space, and damage to the spinal cord results from direct compression, thrombophlebitis of epidural veins, or interruption of the arterial blood supply.
Epidural catheter placement is a common risk factor for SEA (risk is lower when catheters are placed for short time periods). Other risk factors include diabetes mellitus, HIV infection, trauma, tattooing, acupuncture, local spinal injections, and bacteremia secondary to distant infection or intravenous drug use.
The approximate frequency of the bacterial causes of SEA (excluding mycobacteria) is Staphylococcus aureus, 63 percent; Gram negative bacilli, 16 percent; Streptococci, 9 percent; coagulase-negative staphylococci, 3 percent (mostly in patients with prior spinal instrumentation); and anaerobes, 2 percent.
Approximately one-third of patients have no identifiable source for the infection. Among the two-thirds with an identifiable portal of entry, the most common sites are infections of the skin and soft tissues and complications of spinal surgery or other invasive procedures, including epidural catheters left in place for pain control. Up to one–third arise from hematogenous infection. With SEA, the approximate frequency of positive cultures in abscess contents is 90 percent; venous blood, 62 percent; and cerebrospinal fluid (CSF), 19 percent (Gram stain of CSF is usually negative). Surgical decompression and drainage with systemic antibiotic therapy is the treatment for most patients.
The following reference is from UpToDate, Rose BD (Ed), UpToDate, Waltham, MA 2008. Copyright 2008 UpToDate, Inc. Accessed on March 16, 2009. For more information, visit uptodate.com.
Durack DT, Sexton DJ. Epidural abscess.
The guidelines suggested here are not rules, do not constitute legal advice, and do not ensure a successful outcome. The ultimate decision regarding the appropriateness of any treatment must be made by each healthcare provider considering the circumstances of the individual situation and in accordance with the laws of the jurisdiction in which the care is rendered.