Portal de Periódicos da CAPES

Jugular Venous Desaturation Due to Epidural Hematoma After Craniotomy

1996; Lippincott Williams & Wilkins; Volume: 82; Issue: 2 Linguagem: Inglês

10.1097/00000539-199602000-00038

1526-7598

Bernhard Walder, O.H.G. Wilder‐Smith, Alain Reverdin, E. Tassonyi,

Spinal Fractures and Fixation Techniques

The indication for jugular venous bulb catheterization to monitor cerebral oxygen extraction is well established for patients with severe head injuries [1-3]. The use of jugular venous catheters is safe, with a low incidence of complications such as venous thrombosis or increased intracranial pressure [4,5]. For elective craniotomy the indication is less well established, although its usefulness, and the significance of changes, is now being documented [6]. We report a case where jugular venous bulb oxygen saturation (SvjO2) decreased in conjunction with development of a significant acute epidural hemorrhage in the immediate postoperative period. Case Report A 37-yr-old man, weighing 69 kg and 167 cm tall, underwent a craniotomy in the sitting position for the excision of a small left occipital arteriovenous malformation. The patient, who was otherwise in good health, had suffered persistant headache and disturbed vision with right hemianopsia since a hemorrhage 3 mo earlier. Anesthesia was induced and maintained with a continuous intravenous infusion of propofol and alfentanil. No nitrous oxide was used. Intraoperative monitoring included a left radial artery cannula, a right internal jugular central venous catheter, a right retrograde jugular venous bulb catheter, continuous processed electroencephalographic monitoring, and precordial Doppler ultrasound monitoring. The position of the two jugular catheters was established radiographically. During the 3.5-h surgical procedure, the patient remained hemodynamically stable, with mean arterial pressures of 80-100 mm Hg, central venous pressures between 0 and 5 mm Hg and heart rates between 60 and 85 bpm. The SvjO2 was 66% at 30 min after incision (hematocrit 37%, arterial oxygen saturation [SaO2] of 99.7%). Ninety minutes postincision, SvjO2 was 70% (hematocrit 34%, SaO2 99.0%). Intraoperative bleeding was minimal, and no air embolism was detected. Tracheal extubation and recovery from anesthesia were uneventful. In the recovery room, the patient was initially stable from a hemodynamic, respiratory, and neurologic point of view, with a SvjO2 of 59% (hematocrit 36%, SaO2 99.5% with oxygen applied via face mask). The hemianopsia persisted for 3 h postoperatively at which time the patient became restless and bradycardic (47 bpm). He was slightly agitated during neurologic examination and could not remember being in the recovery room. The mean arterial pressure was 100 mm Hg, central venous pressure 3 mm Hg, Glasgow Coma Scale 14, and the pupils were symmetric and reactive to light. The nurse considered his restlessness to be the result of positional discomfort arising from the hard recovery room stretchers, and phoned the attending anesthesiologist to discharge the patient. At this point SvjO2 was 38% (hematocrit 36%, no supplemental oxygen, SaO2 96%). On the basis of the jugular venous oxygen desaturation together with the minor neurologic changes, it was decided to perform urgent cerebral computed tomography which showed a large occipital epidural hematoma which was immediately evacuated. After that intervention, the patient returned to his previous neurologically stable state, with a SvjO2 in the immediate postoperative phase of 62% (with O2 via facemask, SaO2 92.1%, hematocrit 31%). Eleven hours postoperatively, the SvjO2 was 50% without supplemental oxygen (SaO2 98.2%). The patient recovered well, but the disturbances of vision persisted. Discussion The clinical features of an acute postoperative epidural hematoma are relatively nonspecific. Hemiparesis, anisocoria, diminished Glasgow Coma Scale, poor recovery, and a lucid interval have all been described [7]. The development of posterior hemispheric epidural hematomas may be insidious, with usual signs of brain compression and lateralization often being absent. Small epidural hematomas not affecting clinical outcome are frequent (14.4%) [7]. The incidence of clinically relevant postoperative epidural hematomas is 0.9%, with the highest incidence after intracranial hematoma removal [7]. The incidence of clinically relevant postoperative cerebral hematoma at any site is slightly higher (1.1%-2.2%) with most of them occurring within 6 h of surgery [8,9]. Such hematomas are frequent after meningioma surgery (6.2% of all meningioma surgeries), arteriovenous malformation (5.1%), and after craniotomy for trauma (3.7%) [9]. Intracranial pressure (ICP) monitoring has been suggested to enable better and earlier recognition of postoperative cerebral complications after intracranial surgery [10]. Unfortunately ICP monitoring correlates poorly with clinical deterioration after intracranial surgery. In the study by Constantini et al. [10] nearly one third of the patients showed clinical deterioration, but none had an associated increase in ICP. The failure of such monitoring, which can easily produce a false sense of security, may be technical or due to ischemia without concommitant edema. The present case suggests that use of SvjO2 monitoring may be particularly useful in some postoperative neurosurgical patients. If the hemoglobin concentration and the SaO2 remain constant, and if the difference between the arterial and jugular venous partial oxygen pressure is not too large, then SvjO2 will be proportional to the ratio of cerebral blood flow to metabolic rate [11]. Levels less than 54%-45% may indicate global cerebral ischemia [1,3,12]. Jugular venous oxygen desaturation occurs in head-injured patients in the presence of intracranial hypertension, hypocarbia, arterial hypoxemia, persistent systemic hypotension, and cerebral vasospasm [3]. Increases in SvjO2 of more than 75% suggest luxury perfusion or hyperemia. In this case, SvjO2 was not measured continuously. It is conceivable that with a fiberoptic catheter for continuous monitoring, we could have detected the desaturation earlier, thus treating the patient even more promptly, perhaps even before neurologic deterioration occurred. Unfortunately, fiberoptic continuous SvjO (2) remains technically problematic. In a study by Steinberg et al. [3], 50% of SvjO2 desaturations measured with a fiberoptic catheter were false, mainly due to catheter movement. To date, the use of jugular bulb oxygen saturation monitoring after elective intracranial surgery remains little investigated, although its potential benefits have been described in a study of the intermittant intraoperative use of SvjO2 monitoring [6] to detect global cerebral ischemia. As with all invasive monitors, the potential benefits must be balanced against the potential risks (venous obstruction, carotid artery puncture, infection, etc.) in each patient. In conclusion, this case report suggests that SvjO2 monitoring of elective craniotomies may be a promising method for the detection of early postoperative cerebral complications. Further clinical trials are indicated to assess the risks and benefits of routine use of such monitoring for elective intracranial surgery.