| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Section of Pulmonary and Critical Care Medicine, University of Chicago, Chicago, IL.
Correspondence to: John Kress, MD, University of Chicago, Department of Medicine, Section of Pulmonary and Critical Care, 5841 South Maryland Ave, MC6026, Chicago, IL 60637; e-mail: jkress{at}medicine.bsd.uchicago.edu
A 62-year-old man with a history of multiple myeloma was admitted to the medical ICU with respiratory failure due to respiratory muscle weakness. His myeloma had been diagnosed 4 months earlier, when he had presented with a lumbar cord compression and underwent urgent laminectomy. His pathologic specimen revealed myeloma invading the epidural fat. He had undergone 4 months of therapy with dexamethasone and thalidomide, was walking unassisted, and was independent in his activities of daily living. At the time of the present hospital admission, his family described a history of progressive ascending motor weakness over a 6-week period to the point that he was unable to walk.
Physical Examination
On examination, the patient was areflexic bilaterally at his bicep, tricep, brachioradialis, and ankle locations. Patellar responses were hyperreflexic bilaterally. He had symmetric ptosis, marked weakness of forearms and hip flexors, and normal rectal tone. He was cognitively intact. Lung fields were clear bilaterally.
Laboratory and Radiographic Findings
An MRI of the patient’s cervical, thoracic, and lumbar spine showed abnormal enhancing epidural tissue surrounding and flattening the spinal cord at levels C3 through T2, and T8 through L1. This enhancement was interpreted as progressive myelomatous involvement, similar to his presenting lumbar epidural mass 4 months earlier. A lumbar puncture revealed a total protein concentration of 4710 mg/dL (serum protein concentration, 8.5 g/dL) with normal cytology. Electromyography demonstrated an axonal sensorimotor neuropathy of the bilateral radial and tibial nerves, and the right ulnar, peroneal, and sural nerves; in addition, it found evidence for a polyradiculopathy affecting all muscles in the lower limbs, the right thoracic paraspinal muscles, and the right forearm muscles. The study was interpreted as being consistent with neoplastic infiltration of multiple levels of the spinal cord.
Hospital Course
The patient was treated with dexamethasone, 4 mg IV q6h, and IV Ig, but over the subsequent week he experienced increasing weakness, dyspnea, and confusion. Serum viscosity, relative to saline solution, was normal at 1.7. Arterial blood gas levels measured while breathing room air revealed respiratory alkalosis (pH, 7.44; PaCO2, 23 mm Hg; PaO2, 84 mm Hg) with a respiratory rate of 26 breaths/min; his negative inspiratory force was measured at –17 cm H2O. Fourteen days into his hospitalization, the patient was admitted to the ICU with a respiratory rate of 32 breaths/min, increasing somnolence, and the following arterial blood gas levels while breathing 50% O2 by nonrebreather face mask: pH, 7.10; PaCO2, 53 mm Hg; PaO2, 102 mm Hg. His acid-base disorder was complex, with a respiratory acidosis superimposed on a myeloma-induced distal (type I) renal tubular acidosis. He was intubated for ventilatory failure and began receiving mechanical ventilation (model 7200ae; Respironics; Murrysville, PA) in a volume-cycled assist-control mode with the following settings: rate, 14 breaths/min; tidal volume, 450 mL; positive end-expiratory pressure (PEEP), 5 cm H2O; and fraction of inspired oxygen, 50%. Other ventilator settings included a pressure-triggered inspiratory sensitivity of –2 cm H2O, a set flow rate of 60 L/min, and a square waveform. He required no sedation after his induction for intubation and remained passive on the ventilator, with rare spontaneously triggered breaths above his set rate. On these settings, arterial blood gas levels were as follows: pH, 7.27; PaCO2, 36 mm Hg; PaO2, 120 mm Hg. Peak inspiratory pressure was measured at 26 cm H2O, and plateau pressure was measured at 8 cm H2O, which prompted the initiation of nebulized albuterol and ipratropium treatments every 6 h. The patient was a pipe smoker.
On the patient’s fourth day in the medical ICU, physicians were called for "respiratory distress," because the patient was breathing at a rate of 42 breaths/min. He had remained unresponsive despite a lack of sedation and normalization of the arterial PCO2. He was given propofol at a rate of 10 µg/kg/min for his tachypnea, but his respiratory rate remained at 46 breaths/min. His BP fell from 132/64 mm Hg to 79/43 mm Hg; therefore, treatment with vasopressin, 0.04 U/min, was ordered while 2 L of 0.9 normal saline solution were administered IV. On identical ventilator settings, the peak pressure was now 34 cm H2O with a plateau of 18 cm H2O. Auto-PEEP was measured at 7 cm H2O. Chest radiograph findings were unchanged. The arterial blood gas measurements at the initial stage of this tachypnea were as follows: pH, 7.32; PaCO2, 29 mm Hg; PaO2, 109 mm Hg. The ventilator waveform tracings of flow vs time and pressure vs time are shown in Figure 1 .
|
Diagnosis: Ventilator autocycling
Ventilators detect patient effort by sensing either a pressure or flow signal at the interface between the tracheal tube and the ventilator circuit. Depending on the method used, either a modest reduction in pressure (typically –1 or –2 cm H2O below the PEEP) or a slight increase in flow (commonly 1.0 to 2.0 L/min) at the Wye piece is detected, initiating the flow of gas from the ventilator in its preset mode. The phenomenon of inappropriate ventilator auto-triggering due to excess moisture, cardiogenic oscillation, left ventricular assist devices, cuff leaks, and the misplacement of an inline suction catheter has been described. The humidifier has been implicated as the source for moisture that occluded the expiratory limb of the ventilator, after others demonstrated that oscillating water in the expiratory limb could falsely trigger even a model lung that was receiving ventilation. For our patient, we thought that the source of moisture was frequent nebulizer treatments, which created an excess of liquid in his expiratory limb. In contrast to earlier reports describing water occluding the expiratory limb at the pressure transducer, we observed that the liquid pooled in the gravity-dependent portion of the sagging expiratory limb of the ventilator tubing, several feet from the transducer (Fig 2 ).
|
Further Hospital Course
Figure 1 demonstrates that during the time of his tachypnea, the patient had a respiratory rate of 43 breaths/min with the ventilator in a volume-controlled mode with a significant expiratory sawtooth pattern in both the flow and pressure waveform tracings. While such dramatic swings in the expiratory flow are occasionally seen with patient-ventilator asynchrony and ineffective efforts to trigger inspiration, this would be distinctly unusual in a patient such as ours with severe neuromuscular weakness, whom we considered to be too weak to trigger the ventilator. Furthermore, the rapid oscillating nature of the expiratory flow tracing suggests that the source of the problem is external to the patient. An alternative explanation for the sawtooth pattern is the presence of excess liquid in the expiratory limb of the ventilator tubing, which can move back and forth within the tubing in a pendular fashion and can cause small pressure fluctuations. Approximately 120 mL of liquid was visible in our patient’s ventilator tubing expiratory limb, and when this was emptied from the expiratory tubing and the tubing was reconnected, the respiratory rate immediately decreased to the set respiratory rate on the ventilator. His subsequent ventilator waveform, 2 min after emptying the liquid from the expiratory limb of the ventilator tubing, is shown in Figure 3 . It reveals the patient breathing at his set rate, and his auto-PEEP, which was detectable as a persistent expiratory flow on his initial flow waveform, has resolved. The patient’s measured lung mechanics at the time were a peak inspiratory pressure of 26 cm H2O, a plateau pressure of 10 cm H2O, and an auto-PEEP of 1 cm H2O. Within 10 min of rectifying the excess liquid and without other intervention, the patient’s BP had risen to 102/48 mm Hg, underscoring the hemodynamic significance of auto-PEEP. Therapy with vasopressin had been ordered, but was never administered. Repeat arterial blood gas measurements revealed the following levels, which were consistent with his renal tubular acidosis: pH, 7.22; PaCO2, 40 mm Hg; PaO2, 142 mm Hg. Throughout the episode of "respiratory distress," the patient remained unresponsive, without diaphoresis or evidence of any neuromuscular activity.
|
Received for publication October 27, 2005. Accepted for publication November 16, 2005.
Suggested Readings
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
