Ultrasonographic measurement of optic nerve sheath diameter: A point of care test helps in prognostication of Intensive Care Unit patients.

Early identification of elevated intracranial pressure (ICP) is critical to ensuring timely and appropriate management to improve patient outcome. Measurement of the optic nerve sheath diameter by ultrasound is a well studied modality for noninvasive assessment of ICP. Recent studies have shown it to correlate with invasively measured ICP. We utilized this technique in our ICU and found it to be of great help in guiding patient management and predicting the prognosis. A case series of four patients is reported illustrating its utility in ICU patients.

INTRODUCTION

Early diagnosis and prompt treatment of intracranial hypertension is life saving. In intensive care settings, there are clearly barriers to invasive intracranial pressure (ICP) monitoring, which include expense, availability of personnel, patient instability and treatment in non-tertiary care centres.[1] Dependence on remotely located imaging studies to diagnose a cerebral insult requires shifting of patients which is risky and may further delay the diagnosis. Unfortunately, neurological examination and computerised tomography (CT) are not always reliable indicators of ICP status, and well-known findings such as papilledema and pupillary changes often develop very late.[2345] Optic nerve sheath diameter (ONSD) is a surrogate for detecting raised ICP[67] and would be of significant value in the care of the patient when ICP monitoring is unavailable or contraindicated. ONSD in patients with raised ICP above 20 mmHg varies from 0.48[2] to 0.7[8] cm. The learning curve is 25 and 10 procedures for novice and experienced sonographers, respectively,[9] and the average time to perform the ONSD examination is <2 min,[12] with low inter-observer variability (0.01–0.03 cm).[1] The majority of articles on ONSD measurement are in the setting of raised ICP secondary to traumatic brain injury. However, a few studies have used this measurement to diagnose or assess the severity of other pathologies, including meningitis, stroke, hepatic encephalopathy, preeclampsia, epilepsy and acute mountain sickness.[101112131415]

We report a series of four patients in our Intensive Care Unit (ICU) in whom ONSD monitoring not only helped in diagnosing raised ICP but also guided the management and predicted the prognosis.

CASE REPORTS

Case report 1

A 55-year-old male patient presented to the emergency department (ED) with acute intestinal obstruction. He was a chronic smoker, alcoholic, and known case of hypertension, taking antihypertensives for the last 15 years. His biochemical investigations revealed abnormal liver profile (aspartate transaminase [AST], 125 IU; alanine transaminase [ALT] 207 IU; serum alkaline phosphatase, 212 IU; serum. bilirubin = 8mg/dL, international normalised ratio [INR] = 2.0) while rest of the investigations were within normal range. Pre-operative vitals were blood pressure (BP), 160/90 mmHg; heart rate, 128/min; respiratory rate, 34/min; SpO2, 98% (on room air); temperature, 102°F; respiratory and cardiovascular system were normal. The patient was taken for surgery after pre-operative optimisation (hydration, injection paracetamol intravenously and fresh frozen plasma [FFP] transfusion). General anaesthesia was administered using propofol, fentanyl, atracurium, sevoflurane in oxygen and nitrous oxide. The surgery lasted for 5 h during which resection and anastomosis of intestine were performed. Intra-operatively, an intravenous infusion of nitroglycerin 20 mcg/min was started for persistently high BP. The patient was shifted to ICU for observation, following successful extubation. After 6 h, the patient developed respiratory distress (respiratory rate = 44/min); arterial blood gas revealed respiratory alkalosis and hypoxia though the chest was clear to auscultate and Glasgow coma scale (GCS) was 13/15. Despite oxygen therapy, hypoxia did not resolve and elective ventilation was done. ONSD was 0.48 cm in both the eyes. During subsequent days, the GCS further deteriorated (3/10 on the 3rd day), pupils dilated and became sluggishly reactive to light, ONSD increased gradually (0.52 cm on the 2nd day, 0.73 cm on the 3rd day; 40% increase). Liver functions became worse and renal functions were also deranged. were also deranged. Hypertension persisted despite antihypertensive medications. CT scan of the brain revealed large intracerebral haemorrhage. Mannitol and phenytoin were administered intravenously. End-tidal carbon dioxide was maintained between 30 and 35 mmHg. After consultation with the neurosurgeon, decompressive craniotomy and lax duroplasty were done. Post-operatively, there was no improvement in his condition with GCS 3, mydriatic non-reactive pupils and ONSD remained high. The patient succumbed on the 5th post-operative day due to multiorgan failure.

Case report 2

A 24 year old post-partum patient presented to ED with a history of recurrent seizures 6 h after an uneventful vaginal delivery. On examination, her GCS was 5/15, pupils mid-dilated (6mm) and sluggishly reacting to light, SpO2 80% (FiO2 0.4) with BP 186/90. Immediately her airway was secured, elective ventilation was done and loading dose of magnesium sulphate was given. She had history of pregnancy-induced hypertension. Blood investigations revealed haemoglobin concentration of 8.9 g/dl, total leucocyte count, 22,000/cmm; platelets, 60,000/cmm; blood urea, 25 mg/dl; blood sugar, 89 mg/dl; serum sodium, 145 meq/L; potassium, 5 meq/L; serum bilirubin, 8 mg/dl; AST, 92 U/L; ALT, 150 U/L; lactic dehydrogenase, 1240 U/L; INR, 9.5 and serum creatinine, 0.7 mg/dl. Based on the clinical and laboratory data, diagnosis of HELLP (haemolysis, elevated liver enzymes, and low platelet count) syndrome was made. On day 2, other anti-hypertensives were added. FFP and platelet-rich plasma were transfused. The ONSD on the 1st day was 0.62 cm and it increased to 0.72 and 0.79 cm on the 2nd and 3rd day, respectively (20% increase). The GCS remained poor and she developed multiorgan failure. CT scan of the brain revealed intracerebral haemorrhage with 12 mm shift in midline. Mannitol and injection phenytoin were given intravenously. End-tidal carbon dioxide was maintained between 30 and 35 mmHg. Despite best efforts, patient died on the 4th day of admission.

Case report 3

A 19-year-old male patient presented to ED with history of assault. There was history of headache, multiple bouts of vomiting and altered sensorium. The GCS was 7/15, pupils mid-dilated (6mm) and sluggishly reactive to light, tone of both upper and lower limbs was increased and plantars were bilateral silent. Haemodynamically, he was stable but oxygen saturation on air was 60% and auscultation of the chest revealed bilateral crepitations. CT scan of the brain, haematological and biochemical investigations were within normal limits. Tracheal intubation was done and he was put on elective ventilation. The ONSD was 0.60 cm in both eyes. The treatment was started as per standard protocol for head injury. Neurological status remained poor and ONSD remained high (0.67 cm) Figure 1 on the 6th day (11% increase). Magnetic resonance imaging (MRI) of the brain was done which revealed multiple infarcts in watershed areas, periventricular and corticomedullary junction. The conservative management continued and clinical improvement was noticed on the 23rd day; there was decrease in ONSD (0.52 cm). Patient was shifted from ICU on the 35th day and was discharged home on the 42nd day from neurosurgical ward.

Figure 1

Raised ONSD (0.67cm), ocular ultrasound was performed using Sonosite™ M-turbo (SonositeInc, Bothell, WA,USA) ultrasound

Case report 4

A 32 year old man (body mass index 34 kg/m2) presented to the ED with history of road side accident. Patient complained of left-sided chest pain with no history of nausea, vomiting, ENT bleed or seizures. On examination, left-sided pneumothorax was diagnosed and chest drain inserted immediately. Family and medical history were non-contributory. He was a chronic smoker and alcoholic. CT scan of the head was normal. Contrast-enhanced CT of the thorax and abdomen revealed left-sided pleural effusion, pneumomediastinum, splenic, liver and left-sided renal contusion as well as laceration, fractures in the transverse process L1 to L4 vertebrae, 9th–11th ribs at more than two sites, right-sided ilium and iliac crest as well as superior and inferior pubic rami. Patient was shifted to ICU for observation. On examination, GCS was 9/15, patient was irritable, pupils of normal size and reacting to light, SpO2 was 94%. All other vitals were within normal limits. ONSD was 0.56 cm in both the eyes. Patient was put on Venturi mask (FiO2 0.60). All investigations were within normal limits and patient was monitored closely monitoring of all vitals. Deep vein thrombosis prophylaxis with enoxaparin (40 mg OD) was started. On subsequent days, GCS improved to 11/15 with ONSD of 0.5 cm on average. On the 7th ICU day, patient had a GCS of 9/15 not responding to verbal commands, pupils non-reacting to light, SpO2 86%, BP 90/60 mm Hg, heart rate 140/min and temperature 106°F. ONSD was 0.78 cm and 0.79 cm on the right and left side, respectively (41% increase). Raised ONSD along with negative blood, urine and tracheal culture report predicted a central cause of hyperthermia. CT scan of the head repeated and a haemorrhagic cyst compressing the brain (hypothalamus) was diagnosed subsequently. Mannitol and phenytoin were given intravenously and were subsequently intubated; neurosurgical intervention was planned but the patient expired on the 7th day of ICU stay.

DISCUSSION

Ultrasound technology is commonly used in the evaluation of critically ill patients. Focused abdominal sonogram for trauma scan is a simple, non-invasive bedside procedure used largely in place of the more invasive diagnostic peritoneal lavage.[5] Similarly, point of care ultrasonography ONSD acts as surrogate marker for raised intracranial hypertension. The cut-off value of ONSD indicating raised ICP is elusive in the literature. No reports identify a specific value which indicates need for surgery or, alternatively, predicts successful non-operative treatment. ONSD is used only as a screening tool for raised ICP. No cut-off value was determined as an independent indicator of emergent craniotomy.[1] The utility of ONSD as a screening tool for raised ICP during initial trauma patient evaluation is well demonstrated. A cut-off value of 0.5 cm can be used to screen for raised ICP. The sensitivity and specificity to detect ICP ≥20 mmHg were 74%–100% and 63%–95%, respectively, when the cut-off value ranged from 0.50 to 0.59 cm.[2]

In intensive care, most patients are sedated, intubated and sometimes paralysed. During their stay in the ICU, new cerebral insult often goes undetected, adding to the mortality and morbidity of the patients. Sometimes, patients who are difficult to wean from ventilator due to poor GCS are often diagnosed later, having new cerebral insult. To detect raised ICP by clinical examination in such scenarios becomes challenging. Invasive monitoring may not be always available, and shifting of critically ill patients to radiology suite is risky. With the help of ONSD, we could not only detect new cerebral insult but could also monitor the response to medical or surgical intervention done. It helps in defining the prognosis of the patient. In our cases by measuring ONSD, we could not only detect a new cerebral insult (increase of even 11% from baseline ONSD values) but also changed our management accordingly by sending the patient for imaging, starting medications and planning for neurosurgical procedures.

Thus, it may be argued that the main value of this technique lies not only in early evaluation but also during repeated assessment as required where cross-sectional imaging is unavailable. However, more studies are needed to determine if this is a viable strategy to reduce the need for CT or MRI imaging. For patients at risk of raised ICP due to preeclampsia, eclampsia, meningitis, liver failure and stroke, this straightforward bedside assessment may be helpful in risk stratification and decision-making.

CONCLUSION

We suggest ONSD measurement be adopted as a point of care test in ICU patients. It will help detect cerebral insults, monitor neurological status, guide treatment strategy and predict prognosis.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.