On HD 3, the patient’s respiratory status began to deteriorate. Bilateral diffuse alveolar infiltrates developed, hypoxemia worsened, and ventilatory support was increased. The patient’s temperature rose to 38.7°C, and leukocytosis levels increased from 14,000 /μL to 21,800/μL. Broad-spectrum antibiotics, specifically piperacillin/tazobactam and linezolid, were started empirically for ventilator-associated pneumonia but were eventually discontinued after all cultures were confirmed negative. Despite aggressive diuresis of approximately 1.5 L net negative daily, hypoxia persisted, which was noted to temporally correlate with the fluctuation in ventriculostomy output and BP. A representative arterial blood gas measurement during this time revealed a pH of 7.28, Pco2 of 75.6 mm Hg, Pao2 of 68 mm Hg, oxygen saturation of 93% on pressure control ventilation while receiving an Fio2 of 70% and partial end-expiratory pressure of 18 mm Hg. A two-dimensional echocardiogram showed normal left ventricular function and no evidence of left atrial hypertension. On HD 9, a pulmonary artery catheter was inserted; initial results were as follows: pulmonary artery pressure, 50/30 mm Hg (normal, 15-30/6-12 mm Hg); cardiac index, 5.4 L/min/m2 (normal, 2.4-4.0 L/min/m2); systemic vascular resistance, 523 dynes/s/m2/cm5; pulmonary capillary wedge pressure, 9 to 12 mm Hg (normal, 6-12 mm Hg); and pulmonary vascular resistance, 155 to 200 mm Hg (normal, 200-400 dynes/s/m2/cm5). Based on the acute development of hypoxemia and hypertension following a CNS insult, and further supported by a lack of an alternative cause, specifically heart failure, pneumonia, or pulmonary embolism, the diagnosis of neurogenic pulmonary edema (NPE) was made. Despite maximal infusion of nicardipine and multiple other antihypertensive agents including metoprolol, hydralazine, and labetalol, BP remained labile and gas exchange deteriorated in an episodic fashion. Antibiotics and diuresis failed to alter the clinical trajectory. On HD 11, a phentolamine infusion was started at 0.17 mg/min and titrated for BP control. Over 6 h, the Fio2 requirements dropped precipitously, gas exchange improved, and the chest radiograph showed improvement of pulmonary edema (Figs 1, 2). When the hospital supply of phentolamine was exhausted, the clinical status deteriorated rapidly. Within just 15 h of the discontinuation of phentolamine, the Pao2 fell from 166 mm Hg to 66 mm Hg, and Fio2 requirements rose from 60% to 100%. When the phentolamine supply was replenished and the infusion restarted, the same rapid improvement was observed and BP stabilized. Over the next 12 to 24 h, BP also stabilized, allowing for discontinuation of all other antihypertensive agents. The impressive improvement in respiratory parameters over the 3-day span of phentolamine infusion is illustrated in Figure 3. On HD 13, the patient was weaned off phentolamine. Catecholamine levels were measured on three separate occasions, and the results are shown in Table 1.