de intensivist

Abstract

A 50-year-old male patient presented to the emergency department with back and neck pain. During his admission, over a period of a few days, he developed tetraplegia and autonomic dysfunction, after he was initially diagnosed with a misunderstood sepsis and diabetic ketoacidosis. After extensive neurological investigation and additional imaging, the diagnosis of spinal epidural abscess was made. Due to technical reasons, related to the extensiveness of the abscess, surgical intervention was considered impossible. During the weeks that followed, unfortunately, no improvement of the tetraplegia was observed. Due to autonomic dysfunction, mechanical ventilation was indicated and a pacemaker needed to be implanted. Spinal epidural abscess is an uncommon phenomenon and its diagnosis is often delayed due to a lack of awareness. We want to address the importance of early imaging and monitoring neurological functioning in a patient with apparently minor neurological symptoms.

Introduction

Spinal epidural abscess is a rare, but severe infection possibly causing severe and even permanent damage to the nervous system. The prevalence is largest among male patients between 40 and 60 years of age, with an annual incidence increasing over the years, estimated to be 2-3 per 10,000 hospital admissions. The incidence is thought to be increasing due to the higher prevalence of diabetes and the ageing population. Another explanation is that neuroimaging techniques are improving and are becoming widely available.

The major prognostic factor for a favourable clinical outcome is early diagnosis, leading to appropriate treatment. Therefore, prompt recognition is imperative. However, in daily clinical practice, the diagnosis of spinal epidural abscess is often not considered at an early stage, since clear neurological symptoms are frequently absent.

We describe a case of an extensive spinal epidural abscess with a severe clinical disease course after an initial presentation with apparently subtle neurological symptoms.

Case

A 50-year-old male presented to the emergency department with leg cramps, weakness of the right leg, a painful neck and back in addition to profuse sweating for one week. His medical history included obesity (BMI 32.2) and type 2 diabetes mellitus complicated with a foot ulcer, hypertension, obstructive sleep apnoea syndrome and asthma.

Physical examination at presentation showed disturbed vital signs with an oxygen saturation of 87% in room air, a respiratory rate of 35/min and tachycardia of 140/min with normotension (125/85 mmHg). The patient had been unconscious for 10 minutes at arrival, followed by spontaneous regained consciousness with a Glasgow Coma Scale of 15. He showed tremors on both the upper and lower extremities. His body temperature was 37.1 °C.

Laboratory investigation showed high inflammation parameters (leukocytes 33.0 x 10⁹/l and C-reactive protein 350 mg/l) with an acute kidney injury (creatinine 196 µmol/l and urea 18.8 mmol/l). In addition, severe hyperglycaemia was found. Venous blood gas analysis showed a lactic acidosis (lactate 10.1 mmol/l, pH 7.20; bicarbonate 14 mmol/l). Urinalysis showed no signs of infection. Ketones were, however, detected (0.5-5 mmol/l).

Based on these findings the patient was initially admitted to the internal medicine ward with the diagnosis of diabetic ketoacidosis due to sepsis without an apparent focus, but possibly due to the existing foot ulcer. Still, the episode of spontaneous unconsciousness was not understood. During the following hours he became respiratory insufficient and was consequently admitted to the intensive care unit (ICU) for respiratory support.

Because of worsening neck and back pain, an emergency MRI was performed, showing a spinal epidural abscess with lumbar and cervical narrowing of the spinal canal (figures 1 and 2). Consultation of and extensive discussion with an orthopaedic surgeon and several neurosurgeons in different neurosurgical centres led to the conclusion that both surgical intervention and drainage were considered impossible due to the extensiveness of the abscess. The only option was antibiotic treatment for a prolonged period of time. Blood cultures showed the presence of Streptococcus agalactiae. Antibiotic therapy was changed to benzylpenicillin, in accordance to the Dutch guidelines, instead of ceftriaxone which initially had been started.

Due to the respiratory distress the patient was intubated and mechanically ventilated. Two days later, after sedation was stopped, it became apparent that tetraplegia had developed. There was no spontaneous movement of the limbs, no sensitivity in the extremities and no movement of the diaphragm. On day 13 of ICU admission the patient developed an episode of bradycardia and asystole with loss of cardiac output necessitating cardiopulmonary resuscitation. During the following four days this happened several times (four times). We assumed it was the consequence of autonomic neuropathy which led us to the decision to implant an external pacemaker. Due to persistent bradycardia, the external pacemaker was replaced by a permanent pacemaker.

During the days and weeks that followed, unfortunately, no improvement was seen in his neurological status. The patient remained tetraplegic and dependant on the ventilator during his 18-week stay in the ICU. He was ultimately transferred to a nearby revalidation centre.

Figure 1. MRI image of the epidural abscess. Sagittal T2 TIRM of the spine shows an extensive hyperintense epidural collection along the posterior margin of the cord (arrows)

Figure 2. MRI images of the cervical spine
A: Sagittal T2 TIRM shows an extensive hyperintense epidural collection along the posterior margin of the cord (arrows). There is mass effect on the cord with hyperintensity in cord and medulla oblongata (dotted arrow) B: Axial T2 at level C2 (yellow line in figure A) shows an extensive hyperintense epidural collection with mass effect on the cord (dashed arrows)

Discussion

This case report describes a patient with a spinal epidural abscess due to S. agalactiae infection. The problems started with neck and back pain swiftly followed by respiratory failure thought to be the consequence of failure of the diaphragm function, tetraplegia, and even autonomic nervous dysfunction.

Spinal epidural abscess is a rare infection and involvement of the entire spine, known as panspinal involvement, is even rarer.[1,2] It is mostly seen at the thoracic (T) level (50%), followed by lumbar level (34%) and cervical (C) level 15%.[1]

The diagnosis is often recognised at a late stage due to non-specific neurological symptoms and therefore it is often associated with irreversible neurological damage.[2] In the described case the neurological abnormalities found at initial presentation were believed to be part of the diagnosis of ‘diabetic ketoacidosis’. One could suggest that the initial presentation should have warranted a CT scan so that the abscess would have been detected at an earlier stage, although it remains unclear whether that would have led to a different outcome. A classical triad of symptoms is described as back pain (75-85%), fever (32-50%) and a variety of neurological deficits (27-40%).[2-4] This triad, however, is only present in about 8-15% of the cases at first presentation.[2,3] Also, the symptomatology and the course of the disease vary widely. In our case fever was not present.

As has been described before, it is not unusual that a spinal epidural abscess presents as sepsis, with progression into septic shock with multi-organ failure.[5]

A known important predisposing risk factor for the development of spinal epidural abscess is the presence of diabetes mellitus, as in our patient. Other predisposing factors include an abnormality or trauma of the spine, intravenous drug use and a state of immunodeficiency.[2]

One of the most severe complications of spinal epidural abscess is spinal cord injury, believed to be the consequence of compression by the empyema. It causes a broad spectrum of clinical symptoms by disturbing not only sensory and motor functions, but also autonomic neurological function. In our case bradycardia and even asystole developed as an expression of autonomic dysfunction, urging the need for implantation of a pacemaker. Spinal cord injury above the level of T6 is known to be accompanied by cardiovascular complications. Due to disruption of the supraspinal sympathetic pathway, which is located in the cervical cord, acute autonomic imbalances can develop by inducing the parasympathetic dominance via the vagus nerve.[6,7] This presents with severe bradycardia in almost all patients with severe cervical spinal cord injury.

Shortly after hospitalisation, our patient developed respiratory failure. The diaphragm function was found to be impaired on ultrasound, which was explained by the extensive hyperintense epidural collection with a mass effect on the spinal cord at level C2 seen on MRI. Spinal cord injury above the C5 level is typically accompanied by impaired diaphragm function.[6]

Blood cultures showed the presence of group B Streptococcus (GBS). Haematogenous spreading from a distant site or a neighbouring structure is one of the main causes of a spinal epidural abscess. In 30-40% of cases no source of infection can be identified.[1] A streptococcal infection is found in around 9% of cases[2,4] and S. agalactiae is found in only 1.5% of the cases.[2] The main source of GBS infection is the urinary tract. In addition GBS can be found in the lower respiratory tract or in soft tissue infections.[5,6] In our patient, a definite primary source of infection was not found, but it was thought it might have originated from the foot ulcer.

There are no clear guidelines on how to specifically treat spinal epidural abscess secondary to a positive blood culture with a GBS infection. Panspinal involvement is considered a relative, but not absolute, contraindication to surgical intervention, although some studies suggest that the combination of antibiotic treatment and surgical treatment is associated with a better neurological outcome.[5,8] Multidisciplinary consultation led to the conclusion that, due to extensiveness of the spinal involvement, a surgical approach, including drainage of the abscess, was not expected to be useful.

In conclusion in case of unexplained neck or back pain in combination with vague neurological symptoms, a high index of suspicion and early imaging is warranted in order or detect spinal epidural abscess at an early stage.

Disclosures

All authors declare no conflict of interest. No funding or financial support was received.

Informed consent was obtained from the patient for the publication of this case.