Pulmonary alveolar proteinosis (PAP) is a rare pulmonary disease characterized by alveolar accumulation of surfactant. It may be autoimmune in origin or may result from a mutation of surfactant genes, toxic inhalation, and hematological disorders. Autoimmune alveolar proteinosis is the most frequent form of PAP, and whole lung lavage (WLL) remains a mainstay of treatment for autoimmune PAP. There is a need to maintain adequate lung isolation and oxygenation during the lung lavage. We describe the anesthetic challenges faced in 2 such patients and discuss management strategies available to overcome them. Both cases were managed in 2012. Our IRB granted a waiver of consent for this case series.
Case Report 1
A 68-year-old woman was diagnosed with PAP after a transbronchial lung biopsy. She developed a left-sided pneumothorax after the procedure and required a chest tube. Serial chest radiographs after the chest tube insertion showed resolution of the pneumothorax. She was subsequently scheduled for left lung lavage to washout the proteinaceous materials.
Before anesthetic induction, her oxygen saturation was 74% on room air with a PaO2 of 39 mm Hg. An 18-G IV catheter, 22-G arterial catheter, and 7-Fr central venous catheter were inserted. In addition to standard anesthetic monitoring, bispectral index and temperature were monitored. The left-sided chest tube was oscillating and draining well.
After 5 minutes of oxygen administration, her oxygen saturation increased to 100%. Anesthesia was induced with 2 mg/kg propofol, 1.5 μg/kg fentanyl, and 0.4 mg/kg rocuronium. An Arndt endobronchial blocker (Cook Medical, Bloomington, IN) was inserted, using the loop technique, followed by a 35-Fr left-sided double-lumen endotracheal tube (DLT) that sits alongside the endobronchial blocker. Of note, the Seldinger technique can be used to place the endobronchial blocker.a The DLT was inserted until the bronchial lumen was in the trachea and mechanical ventilation started. A fiberoptic bronchoscope was inserted through the bronchial lumen into the loop at the distal end of the bronchial blocker. The bronchial blocker was then guided to isolate the posterior and lateral segments of the left lower lobe (Fig. 1A). These segments were identified by the evidence of the recent transbronchial biopsy and were deemed to be the cause of the left pneumothorax. The bronchial lumen was subsequently advanced into the left main bronchus (Fig. 1B). The patient’s anesthesia was maintained with desflurane using pressure-controlled ventilation with 100% oxygen. Oxygen saturation remained 99% to 100% throughout induction.
Left lung lavage was started with the patient in the supine position, as the right lung was ventilated and the posterior and lateral segments of the left lower lobe isolated. Each cycle involved instilling 400 to 500 mL warm normal saline into the lung via a Y-connector with the drainage port clamped and subsequent drainage via the other limb of the Y-connector. The effluent was drained passively into a bottle placed below the patient, and this was assisted by chest percussion. A total of 16.1 L was instilled. The chest drain in situ was also closely monitored for a sudden increase in drainage in the event of hydropneumothorax. Oxygen saturation throughout lung lavage was between 94% and 100% on FIO2 1.0.
At the end of the procedure, 2-lung ventilation was restored, and the DLT was exchanged for a size 8 endotracheal tube. The patient was sent to the intensive care unit (ICU) for postoperative monitoring. A postoperative chest radiograph revealed a reduction in midzone and lower zone air space opacification with no obvious fluid in the pleural space, hydropneumothorax, or worsening of the pneumothorax.
The patient returned for right lung lavage 2 days later. A similar approach was used apart from the endobronchial blocker because there was no pneumothorax in the right lung. She was readmitted to the ICU after the procedure for postoperative monitoring where her trachea was successfully extubated the next day.
Case Report 2
Our second patient was a 56-year-old woman diagnosed with PAP since 1993. She had undergone 4 previous WLLs. A computed tomography scan of the thorax showed scattered areas of ground glass opacities bilaterally with a background of subpleural pulmonary fibrosis and traction bronchiectasis in both lower lobes, which appeared to be worse on the right side. She was scheduled for right lung lavage.
Before induction, a 20-G IV catheter and a 20-G arterial catheter were inserted. The patient’s anesthesia was induced with a combination of 2.5 mg/kg propofol, 2 μg/kg fentanyl, and 0.6 mg/kg rocuronium. She was then tracheally intubated with a 32-Fr DLT and maintained on desflurane.
With the patient in the supine position, 1.5 L warmed saline was introduced into the right lung in the first washout. However, the oxygen saturation decreased to 62%, and there was a significant reduction in the compliance of the left lung. In addition, only 700 mL of effluent returned after the first washout. The left lung was immediately examined with the bronchoscope and found not to be contaminated. She was then placed on 100% oxygen and her lungs manually ventilated until her saturations increased to 98%. The next few aliquots of saline were limited to 300 mL but she tolerated each washout poorly with frequent desaturations to 80%.
The anesthetic team decided to turn the patient to the left lateral position in an attempt to improve her ventilation–perfusion matching. In the left lateral position, the patient was able to tolerate the washout better with less desaturation. Continuous positive airway pressure ventilation was also applied to the nonventilated lung between lavages. The lowest recorded oxygen saturation in the left lateral position was 86%.
A total of 8 L of warmed saline was introduced, and 7.3 L of saline was retrieved. The patient was also actively warmed throughout the operation.
At the end of operation, the DLT was exchanged with a size 7.5 endotracheal tube, and she was sent to the ICU for postoperative monitoring.
The patient returned 1 month later for left lung lavage. The same anesthetic team was present for the second lavage. However, total IV anesthesia was chosen instead. The patient’s anesthesia was induced with target-controlled infusion of 2.5 μg/mL propofol at effect-site concentration (Schneider model) and 1 ng/mL remifentanil (Minto model).1,2 Paralysis was achieved with 0.6 mg/kg rocuronium.
She was placed on her right lateral side. She remained stable throughout the washout without desaturations. A total of 8 L saline was introduced and 7.05 L retrieved. The patient was sent to the ICU after exchange with a single-lumen endotracheal tube. She was tracheally extubated on the same day and was discharged the next day.
PAP is a rare pulmonary disease with a prevalence of approximately 3.7 cases per million per year.3,4 Patients commonly present with cough and dyspnea and less commonly chest pain and hemoptysis. Although many therapeutic approaches have been suggested, WLL remains the mainstay therapy.5
Common complications associated with WLL include hypoxemia, hemodynamic instability, hypothermia, and hydro- or pneumothorax. There are many reasons why hypoxemia can occur. One of the main reasons is that the technique requires the use of one-lung ventilation on a patient whose pulmonary function is already compromised. Throughout the procedure, there is always a risk of the lavage fluid spilling over to the ventilated lung because of the displacement of the DLT. Moreover, ventilation–perfusion mismatch will occur, especially during the drainage phase when shunt increases. Hemodynamic instability may also occur when a large amount of lavage fluid results in mediastinal shift affecting the filling and ejection of the heart.6
The anesthesia technique described in the literature commonly involves the use of a DLT to achieve lung isolation. The more affected lung is usually lavaged first and if both lungs are equally affected, the left lung, given its smaller volume, is lavaged first. For the first case, we decided to lavage the left lung first, especially in the presence of the pre-existing pneumothorax that compromised respiratory function.
With regard to positioning, supine and lateral decubitus positions have been used. Washing the dependent lung in a lateral decubitus position will prevent flooding of the nondependent lung, but this risks a higher ventilation–perfusion mismatch.7,8 We adopted the supine position in the first case and only tilted the patient slightly during chest percussions. In the second case, we turned the patient so that the ventilated lung was dependent, which resulted in improved ventilation–perfusion matching and oxygenation (Fig. 2).
Displacement of a DLT can lead to the loss of lung isolation. Spillage of lavage fluid can result in severe bronchospasm.6 Nandkumar et al.8 suggested 3 methods to detect the loss of lung isolation: the appearance of bubbles in the lavage fluid draining from the lavage side; an increase in resistance to ventilation on the ventilated side; and/or a decrease in arterial oxygen saturation and increase in end-tidal carbon dioxide that signals insufficient ventilation.
Other methods to treat hypoxemia have been suggested. El-Dawlatly et al.9 described the successful use of continuous positive airway pressure on the lavaged lung during the drainage phase of WLL as demonstrated in our second patient (Fig. 2).
Another important consideration is the phenomenon of hypoxic pulmonary vasoconstriction, whereby blood is shunted to the portion of the lung that is better ventilated in response to hypoxia. Volatile anesthetic agents inhibit hypoxic pulmonary vasoconstriction in a dose-dependent manner, and this will affect oxygenation in patients with underlying lung pathology or having one-lung ventilation.10 Total IV anesthesia may be a reasonable alternative to provide anesthesia and oxygenation in situations in which there is a need to create a large shunt fraction in patients with poor lung function, as demonstrated during the second lavage for our second case report.
Preventing hypothermia improves patient outcome. For our patients, heat was preserved by the use of a warming mattress and warm saline for lavage. Temperature was monitored throughout the procedure.
Our first case was a challenging situation in which a pneumothorax further complicated anesthetic management. Warm saline could have potentially escaped into the pleural cavity and into the chest drain when we did the left lung lavage. We would therefore be unable to wash out the lipoproteinaceous material from the left lung. There was a need to prevent the pneumothorax from happening. By placing the endobronchial blocker in the segment where the transbronchial biopsy was done, we were able to isolate the source of the pneumothorax.
Anesthesia for WLL remains a challenge for anesthetists today. Hypoxemia is commonly encountered. We reported 2 patients with PAP who successfully underwent WLL despite the difficulties encountered. A multidisciplinary team approach with careful planning plays an important role in such challenging situations.
The authors thank Mr. Evan Hock-Tat Lim from SingHealth Academy for the medical illustration.
1. Levitt DG, Schnider TW. Human physiologically based pharmacokinetic model for propofol. BMC Anesthesiol. 2005;5:4
2. Minto CF, Schnider TW, Egan TD, Youngs E, Lemmens HJ, Gambus PL, Billard V, Hoke JF, Moore KH, Hermann DJ, Muir KT, Mandema JW, Shafer SL. Influence of age and gender on the pharmacokinetics and pharmacodynamics of remifentanil. I. Model development. Anesthesiology. 1997;86:10–23
3. Borie R, Danel C, Debray MP, Taille C, Dombret MC, Aubier M, Epaud R, Crestani B. Pulmonary alveolar proteinosis. Eur Respir Rev. 2011;20:98–107
4. Davis KR, Vadakkan DT, Krishnakumar EV, Anas AM. Serial bronchoscopic lung lavage in pulmonary alveolar proteinosis under local anesthesia. Lung India. 2015;32:162–4
5. Chan AK, Takano A, Hsu AL, Low SY. Pulmonary alveolar proteinosis. J Thorac Dis. 2014;6:E77–80
6. Rebelo HM, Guedes L, Veiga D, Fiuza AC, Abelha F. Anaesthetic, procedure and complications management of serial whole-lung lavage in an obese patient with pulmonary alveolar proteinosis: case report. Rev Bras Anestesiol. 2012;62:869–77
7. Webb ST, Evans AJ, Varley AJ, Klein AA. Anaesthesia for serial whole-lung lavage in a patient with severe pulmonary alveolar proteinosis: a case report. J Med Case Rep. 2008;2:360
8. Nandkumar S, Desai M, Butani M, Udwadia Z. Pulmonary alveolar proteinosis with respiratory failure—anaesthetic management of whole lung lavage. Indian J Anaesth. 2009;53:362–6
9. El-Dawlatly A, Hajjar W, Alnassar S, Alsafar R, Abodonya A. Continuous positive airway pressure ventilation during whole lung lavage for treatment of alveolar proteinosis—a case report and review of literature. Saudi J Anaesth. 2011;5:76–8
10. Lumb AB, Slinger P. Hypoxic pulmonary vasoconstriction: physiology and anesthetic implications. Anesthesiology. 2015;122:932–46