Pulmonary alveolar proteinosis

A Case Report and Review

Pulmonary alveolar proteinosis is characterized by the accumulation of proteinaceous material in the alveoli leading to varying degrees of impairment in gas exchange. Generally, the degree of hypoxemia is mild, and it is rare to have respiratory failure requir ing mechanical ventilation. We present a 53-year-old woman with the most severe degree of hypoxemia associated with alveolar proteinosis reported in the English-language adult literature. Her therapy of sequential whole-lung lavage performed while receiving venovenous extracorporeal membrane oxygenation in one operative session is the first reported successful use of this approach. (CHEST 2001; 120:1024-1026)

Key words: extracorporeal membrane oxygenation; hypoxemia; lung lavage; pulmonary alveolar proteinosis

Abbreviations: ECMO = extracorporeal membrane oxygenation; [FIO.sub.2] = fraction of inspired oxygen; PAS = periodic acid Shiff; WLL = whole-lung lavage

Pulmonary alveolar proteinosis represents a rare syndrome characterized by the accumulation of large amounts of proteinaceous phospholipid-laden material in the alveoli. This leads to impaired gas exchange and arterial hypoxemia of varying degrees. Whole-lung lavage (WLL), first described in 1963,[1] remains the most effective treatment. However, this technique can be difficult and dangerous to perform in the most severely hypoxemic patients.

We report the case of a woman with, to our knowledge, the most severe degree of refractory hypoxemia associated with alveolar proteinosis reported in the literature. WLL was performed sequentially while she received venovenous extracorporeal membrane oxygenation (ECMO) in one operative session. In previously reported cases utilizing venovenous ECMO, only one lung was lavaged during an operative session, and the patients remained on ECMO for [is greater than] 24 h.[2-4]

CASE REPORT

A 53-year-old woman presented to an outside hospital with progressive dyspnea, general malaise, and a decreased appetite over a duration of approximately 3 weeks. Chest radiography revealed bilateral alveolar infiltrates. ECG findings were remarkable for only sinus tachycardia. Abnormal laboratory findings at presentation included the following arterial blood gas measuremen(s: pH, 7.,5; [PCO.sub.2], 30 mm Hg; Pa[O.sub.2], 30 mm Hg on room air; hemoglobin, 18.7 g/dL; and hematocrit, 55 %. An initial diagnosis of congestive heart failure and/or pneumonia was made; the patient was treated accordingly with IV diuretics and broadspectrum antibiotics, and was admitted to the ICU. Although she had a good response to diuretics, her dyspnea and hypoxemia tidied to improve and she was subsequently intubated and administered mechanical ventilation, with fraction of inspired oxygen [(FIO.sub.2)] of 1.0.

Her medical history included previous atrial fibrillation treated with procainamide, and gastroesophageal reflux and osteoarthritis. Medications at presentation were lansoprazole, conjugated estrogens, and alprazolam. She had a 60-pack-year smoking history and was a recovering alcoholic.

The patient underwent bronchoscopy with BAL and transbrontrial biopsies without complication. Milky fluid was obtained from lavage of the right middle lobe, and pathologic examination showed pink staining periodic acid-Schiff (PAS)-positive material. Biopsy specimens showed preserved alveolar architecture, again with a PAS-positive staining extracellular substance consistent with pulmonary alveolar proteinosis (Fig 1). The slides were sent to a consulting pathologist who concurred with the initial reading. There was no growth of bacteria or fungi from the lavage specimen and biopsy, both initially and at 6 weeks. The patient was extubated after the procedure but remained on 100% oxygen by mask. She was sent to our institution for WLL.

[ILLUSTRATION OMITTED]

The patient arrived at our institution 7 days after initial presentation. Arterial blood gas analysis measurements at that time were as follows: pH, 7.5; [PCO.sub.2], 39 mm Hg; and Pa[O.sub.2], 68 mm Hg on a 100% nonrebreather face mask. Physical examination revealed bilateral, fine inspiratory rales throughout all lung fields. There was no digital clubbing. A high-resolution CT scan of the chest was obtained and showed the classic findings of widespread airspace consolidation with thickened interlobular septa producing a "crazy paving" pattern (Fig 2).

[ILLUSTRATION OMITTED]

Due to the patient's severe hypoxemia, WLL was performed 36 h after transfer using venovenous ECMO. The patient was reintubated on the evening prior to the procedure because of worsened hypoxemia: while receiving 100% oxygen via nonrebreather face mask, her Pa[O.sub.2] was 48 mm Hg. The following morning, immediately prior to the procedure, results of arterial blood gas analysis while receiving mechanical ventilation at Floe of 1.0 and positive end-expiratory pressure of 8 mm Hg were as follows: pH, 7.36; [PCO.sub.2], 51 mm Hg; and Pa[O.sub.2], 54 mm Hg.

The patient was brought to the operating room, and general anesthesia was initiated. The existing single-lumen endotracheal tube was exchanged for a double-lumen tube in order to selectively ventilate and lavage each lung independently. Tube placement was verified with bronchoscopic visualization. The cardiovascular surgical team then inserted the catheters for ECMO. A 22F catheter was inserted percutaneously into the right internal jugular vein for return of oxygenated blood, and a 28F catheter was directly inserted into the right femoral vein for blood flow into the oxygenator/pump setup. After 10 min on ECMO, while still receiving [FIO.sub.2] of 1.0, her Pa[O.sub.2] increased to 299 mm Hg. However, with only a single lung being ventilated on ECMO, her Pa[O.sub.2] was 83 mm Hg and oxygen saturation was 97%.

The right lung was lavaged first, with the left lung being independently ventilated. We used warmed saline solution contained within normal bladder irrigation bags hung at 30 cm above the midchest. We utilized the tyical Y-type setup with 1/4-inch perfusion "prime-line" tubing. Using this larger-diameter tubing, we were able to complete the entire lavage in nearly half the time a procedure utilizing typical tubing would have taken. A running balance of fluid lavaged and returned was carefully maintained. After receiving ventilation with [FIO.sub.2] of 1.0, each lung was degassed for 5 min prior to the infusion of fluid. A total of 13.8 L was instilled in 10 exchanges, and 11.75 L returned from the right lung, requiring 94 min to complete. Prior to lavaging the left lung, both lungs were ventilated for approximately 10 min. The left lung was lavaged with a total of 11.2 L of fluid in 10 exchanges with a return of 11.0 L, requiring 68 min to complete. Total time on ECMO was 203 min.

Overall, the patient tolerated the procedure quite well. As expected, oxygenation improved slightly during instillation of the fluid and worsened slightly during drainage. At no time, however, did the oxygen saturation drop to [is less than] 86%. Of more clinical significance was the predictable phenomenon of hypotension during instillation of fluid followed by a rapid return of BP during drainage. Generally, there was an initial drop of approximately 40 points in the systolic pressure at the start of instillation. This required the administration of vasopressor support that was discontinued at the start of drainage.

After completion of WLL, ECMO was discontinued, the cannulas were removed, and the patient was transferred directly back to the medical ICU with a single-lumen endotracheal tube. Initial [PO.sub.2] when ECMO was discontinued was 65 mm Hg on [FIO.sub.2] of 0.8. Two hours postprocedure, the [PO.sub.2] was 80 mm Hg on [FIO.sub.2] of 0.6. At 16-h postprocedure, the patient had been weaned to an [FIO.sub.2] of 0.40 with a [PO.sub.2] of 87 mm Hg. She was extubated 30 min later, and had a postextubation Pa[O.sub.2] of 66 mm Hg on 40% oxygen by face mask. Her dyspnea had markedly improved, and she stated that her "lungs felt much lighter." She continued to do well and was discharged the following day receiving home oxygen at 3 L/min by nasal cannula. Of note, her chest radiograph showed only minimal improvement in infiltrates at discharge.

When seen in our office 10 days later, her dyspnea had markedly improved although she still felt "tired." Her resting [PO.sub.2] on room air was 90 mm Hg. Two weeks after this visit, the patient was readmitted for progressive dyspnea and hypoxemia. She underwent repeat lavage of each lung on consecutive days performed in the standard fashion without ECMO. After discharge, she rapidly improved and was taken off supplemental oxygen 1 week later. Nine months after initial lavage, she is no longer hypoxemic and her exercise tolerance has returned to near baseline.

DISCUSSION

In this report, we describe the most severely hypoxemic adult patient with respiratory failure secondary to pulmonary alveolar proteinosis ever reported in the literature. This also represents the first case in an adult where bilateral sequential WLL was performed under venovenous ECMO with discontinuation of bypass prior to leaving the operating room.

The main physiologic disturbance in this disease is impairment of gas exchange secondary to filling of the alveoli with phospholipoproteinaceous material rich in surfactant proteins. This leads to varying degrees of hypoxemia secondary to mismatching of ventilation and perfusion. Generally, patients are only mildly hypoxemic, and it is rare to present with such severe refractory hypoxemia as our patient. It is also extremely uncommon for patients to require mechanical ventilation. In the largest case series in the English-language literature (34 patients), the mean Pa[O.sub.2] prior to lavage was 63 mm Hg with a range of 38 to 80 mm Hg.[5] In another case series of 23 patients, the mean Pa[O.sub.2] was 56 mm Hg, with only two patients having a Pa[O.sub.2] [is less than] 40, the lowest being 31 mm Hg.[6] In all cases of patients treated at our own institution, we have not seen nearly this degree of hypoxemia, nor have we ever had to use ECMO to perform WLL.

In the adult population, there are a total of three reported cases utilizing venovenous ECMO,[2-4] and two reported cases utilizing veno-arterial ECMO.[7,8] In all five eases, only one lung was lavaged in each procedure. In each of the three cases utilizing venovenous bypass, patients remained on ECMO for extended periods between each lung lavaged, as follows: (1) each lung separated by 12 to 52 h total while receiving ECMO,[2] (2) each separated by 12 to 32 h total,[3] and (3) three separate procedures over the course of 6 days receiving ECMO.[4]

During the lavage procedure, blood is diverted to the ventilated lung when the lung undergoing lavage fills with fluid, and returns to the lavage lung when undergoing drainage.[9] This is the basis for the hypoxemia commonly seen during WLL, and the reason for concern when a patient has refractory hypoxemia even prior to the procedure. An alternative to ECMO, which has been reported only in two cases for performing WLL in severely hypoxemic patients, is to use hyperbaric oxygen conditions in a large walk-in chamber.[10] Oxygenation markedly improved in these two patients. Although this technique is less invasive than ECMO, a chamber of the size needed would not be available in most areas. Another modality, which has been tried with poor success, is to perform lobar lavages via fiberoptic bronchoscopy at separate times to avoid the hypoxemia associated with WLL.[11] In general, this technique remains a poor choice as an initial treatment for a patient with hypoxemie respiratory failure.

The utilization of ECMO in our patient allowed us to safely perform WLL without life-threatening hypoxemia at any point. The benefits of WLL clearly outweigh the risks associated with initiation of ECMO. It is our feeling that lavaging both lungs in one session enabled us to immediately discontinue partial bypass at the end of the procedure. It is possible that in the other reported cases utilizing ECMO, patients remained on bypass for prolonged periods because of performing only single lung lavage. In cases of severe alveolar proteinosis where ECMO is utilized, every effort should be made to lavage both lungs in one session.

REFERENCES

[1] Ramirez RJ, Schultz RB, Dutton RE. Pulmonary alveolar proteinosis: a new technique and rationale for treatment. Arch Intern Med 1963; 112:419-431

[2] Cooper JD, Duffin J, Glynn MFX, et al. Combination of membrane oxygenator support and pulmonary lavage for acute respiratory failure. J Thorac Cardiovasc Surg 1976; 71:304-308

[3] Zapol WM, Wilson R, Hales C, et al. Venovenous bypass with a membrane lung to support bilateral lung lavage. JAMA 1984; 251:3269-3271

[4] Sivitanidas E, Tosson R, Wiebalck A, et al. Combination of extracorporeal membrane oxygenation (ECMO) and pulmonary lavage in a patient with pulmonary alveolar proteinosis. Eur J Cardiothorac Surg 1999; 15:370-372

[5] Prakash US, Barham SS, Carpenter HA, et al. Pulmonary alveolar phospholipoproteinosis: experience with 34 cases and a review. Mayo Clin Proc 1987; 62:499-518

[6] Kariman K, Kylstra A, Spock A. Pulmonary alveolar proteinosis: prospective clinical experience in 23 patients for 15 years. Lung 1984; 162:223-231

[7] Altose MD, Hicks RE, Edwards MW Jr. Extracorporeal membrane oxygenation during bronchopulmonary lavage. Arch Surg 1976; 111:1148-1153

[8] Freedman AP, Pelias A, Johnston RF, et al. Alveolar proteinosis lung lavage using partial cardiopulmonary bypass. Thorax 1981; 36:543-,545

[9] Wasserman K, Blank N, Fletcher G. Lung lavage (alveolar washing) in alveolar proteinosis. Am J Med 1968; 44:611-617

[10] Jansen HM, Zuurmond WWA, Roos CM, et al. Whole lung lavage under hyperbaric oxygen conditions for alveolar proteinosis with respiratory failure. Chest 1987; 91:829-832

[11] Brach BB, Harrell JH, Moser KM. Alveolar proteinosis lobar lavage by fiberoptic bronchoscopic technique. Chest 1976; 69:242-247

(*) From the Section of Pulmonary and Critical Care Medicine, Rush-Presbyterian-St. Luke's Medical Center, Chicago, IL. Manuscript received September 7, 2000; revision accepted February 21, 2001.

Correspondence to: Michael R. Silver, MD, FCCP, Section of Pulmonary and Critical Care Medicine, Rush-Presbyterian-St. Luke's Medical Center, 1753 W. Congress Parkway, Chicago, IL 60612; e-mail: msilver2@rush.edu

COPYRIGHT 2001 American College of Chest Physicians
COPYRIGHT 2001 Gale Group