Hypoxia from vasculitic pulmonary haemorrhage improved by prone position ventilation

C. Hayes-Bradley

Anaesthetics Department, Royal Free Hospital, Pond Street, London NW3 2QG, UKE-mail: clarehayesbradley@yahoo.com

Accepted for publication: December 12, 2003


    Abstract
 Top
 Abstract
 Introduction
 Case report
 Discussion
 References
 
Ventilation in the prone position is used in patients with acute respiratory distress syndrome (ARDS), although data supporting this strategy are limited, and benefit for patients with other conditions is unclear. The patient in this report had severe hypoxaemia from diffuse alveolar haemorrhage caused by vasculitis with positive antineutrophil cytoplasmic antibodies (ANCA). Ventilation in the prone position improved oxygenation dramatically. This improvement was initially maintained when returned supine, accompanied by increased ventilation. Prone ventilation was used on three consecutive days for 10, 14 and 15 h, respectively. Prone ventilation could improve oxygenation by better ventilation–perfusion (V/Q) matching and improved drainage of blood from the dorsal lung. The improved oxygenation in this patient should encourage the use of prone ventilation in other patients with pulmonary haemorrhage and severe hypoxia.

Br J Anaesth 2004; 92: 754–7

Keywords: blood, haemoptysis; complications, pulmonary haemorrhage; complications, vasculitis; position, prone; ventilation, oxygenation


    Introduction
 Top
 Abstract
 Introduction
 Case report
 Discussion
 References
 
I report a 68-yr-old man with respiratory failure and diffuse alveolar haemorrhage from vasculitis with positive antineutrophil cytoplasmic antibodies (ANCA). Hypoxaemia was resistant to conventional ventilatory measures, and ventilation in the prone position was started. This improved oxygenation and ventilation dramatically. The reasons for this improvement are considered, as well as other possible treatment strategies for vasculitis and pulmonary haemorrhage.


    Case report
 Top
 Abstract
 Introduction
 Case report
 Discussion
 References
 
A 68-yr-old white male presented with a 1 week history of increasing spontaneous bright red haemoptysis and dyspnoea. An ex-smoker, he had no previous respiratory problems, and denied any associated symptoms. He had had a single minor episode of rectal bleeding one month before, which had not recurred. Other than hypertension and prostatic hypertrophy for which he took nifedipine, aspirin, and tamsulosin he was usually well. He had had a gastric ulcer 20 yr before. Initial examination revealed diffuse bilateral crepitations in his chest and a purpuric skin rash over both legs. Assessment of the upper respiratory tract, central nervous system and abdomen was unremarkable. Urine testing showed 4+ protein and 2+ blood. He was anaemic with a haemoglobin concentration of 8.0 g dl–1 and a white cell count of 22 x 109 litre–1. Coagulation screen and platelet count were normal. Plasma creatinine was 174 µmol litre–1. A chest radiograph showed diffuse perihilar infiltrates (Fig. 1). An initial diagnosis of a pulmonary-renal syndrome was made, and he was given high concentration oxygen and methylprednisolone 500 mg i.v.



View larger version (113K):
[in this window]
[in a new window]
 
Fig 1 Admission chest radiograph showing bilateral diffuse pulmonary infiltrates consistent with vasculitic pulmonary haemorrhage.

 
Over the next few hours haemoptysis and hypoxia worsened (PaO2 6 kPa when breathing air) necessitating tracheal intubation and controlled ventilation. No upper respiratory abnormalities were noted at laryngoscopy. He was given pancuronium, morphine, and midazolam infusions. Three units of packed cells were transfused and cyclophosphamide 150 mg daily was given. Coagulation tests were normal but copious blood stained tracheal aspirates were obtained. Next day, serology confirmed a strongly c-ANCA positive vasculitis (PR3 titre 452), and plasmapheresis was started with 4 litres exchanged for 3 litres colloid (etherified starch) and 3 units of fresh frozen plasma.

Despite one plasma exchange and cyclophosphamide, gas exchange worsened. Ventilation was with pressure control mode, FIO2 1.0, positive end-expiratory pressure (PEEP) 18 cm H2O and an inspiration:expiration ratio of 2:1. Oxygen saturations were 70–80% (PaO2 6.3 kPa), and tracheal suction gave only brief improvement. On turning into the prone position, copious, heavily bloodied tracheal secretions were obtained and oxygen saturation improved (Fig. 2). In the prone position, minute ventilation (MV) increased from 9.2 litre min–1 to 13.2 litre min–1, and PaCO2 decreased from 9.1 kPa to 7.0 kPa. With airway pressure set at 22 cm H2O, tidal volume increased from 420 ml to 590 ml. After 10 h he was returned to the supine position and the improvement in oxygenation was initially maintained (Fig. 2). MV remained increased at 14.1 litre min–1 and PaCO2 6.4 kPa. On the two subsequent days he was placed in the prone position for 14 and 15 h, respectively because of further impairment of oxygenation. Similar but less dramatic improvements occurred. No complications were encountered.



View larger version (15K):
[in this window]
[in a new window]
 
Fig 2 Improvements in FIO2 and PaO2:FIO2 ratios seen with prone ventilation. The decrease pf positive end-expiratory pressure (PEEP) after the second and third prone episodes is shown.

 
Plasma exchange was repeated daily for 4 days. Steroid therapy and cyclophosphamide were continued. He was given broad-spectrum antibiotics, though no infection was detected. Renal function worsened, and plasma urea and creatinine concentrations increased (from 15.3 mmol litre–1 and 173 µmol litre–1 on day 1, to 30.7 mmol litre–1 and 316 µmol litre–1 on day 5, respectively). Haemofiltration was started, with alprostadil infusion to prevent filter clotting, changing to heparin infusion on day 6. No increase in tracheal aspirates was noted with the anticoagulation.

Ventilation was weaned as his respiratory function improved, but he remained oliguric and needed haemofiltration. After removal of the tracheal tube (day 10) he was transferred to the renal unit for further management.


    Discussion
 Top
 Abstract
 Introduction
 Case report
 Discussion
 References
 
Treatment for hypoxia from pulmonary haemorrhage caused by vasculitis can be considered as follows—management of the vasculitis, management of the lung haemorrhage, and treatment of the hypoxia. Ventilation in the prone position has been mostly studied in patients with acute respiratory distress syndrome (ARDS), but may be useful to treat the hypoxia of pulmonary haemorrhage.

The ANCA-associated vasculitides are syndromes previously subdivided into Wegener’s, Churg Strauss and microscopic polyangiitis.1 Pulmonary haemorrhage occurs from alveolar capillaritis.1 2 Alveolar capillaries rupture into alveoli, rapidly filling them with blood.1 If patients survive the initial massive haemoptysis and respiratory failure they may secondarily develop pulmonary fibrosis or ARDS.1 At presentation, patients with vasculitic pulmonary haemorrhage may fulfil ARDS criteria, however, haemorrhage into alveoli is a specific pathology, not common to all ARDS patients, warranting separate consideration.

Management of the vasculitis includes immunosuppression and plasma exchange. Methylprednisolone and cyclophosphamide are well established therapies (sometimes with prophylaxis for cyclophosphamide-induced haemorrhagic cystitis). Pulmonary haemorrhage typically stops within 3–7 days of starting steroid therapy, although it can last longer.1 Plasma exchange/pheresis in ANCA-positive vasculitides is as yet unproven but often used. Evidence for improved outcome with plasma exchange in vasculitic pulmonary haemorrhage comes from Goodpasture’s syndrome (a similar clinical picture to the ANCA-positive vasculitides but caused by circulating anti-basement membrane antibodies). Other treatment considerations are intravenous immunoglobulin and prophylaxis against Pneumocystis carinii pneumonia.1

Serious pulmonary haemorrhage from causes other than trauma has a mortality of 30–85%.3 Identification and tamponade of bleeding from a large vessel by flexible/rigid bronchoscopy may be possible, or a bleeding lung can be isolated by double lumen endotracheal tube.4 A single bleeding source can then be embolized. Intrabronchial iced saline lavage, epinephrine and tranexamic acid can be effective with a localized bleeding source, either for temporary or permanent relief,4 but are not useful in diffuse haemorrhage.

Diffuse lung haemorrhage from vasculitis may respond to activated factor VII.5 Despite normal coagulation studies, patients with systemic vasculitis can have increased tissue factor pathway inhibition, which can inhibit endogenous tissue factor/VIIa complex. By giving factor VIIa, this inhibition is bypassed and vessel coagulation improves. One such report of factor VIIa reducing vasculitic pulmonary haemorrhage has been published.5 Systemic desmopressin and tranexamic acid can also be considered.

In refractory hypoxia, inhaled nitric oxide can selectively vasodilate ventilated lung and improve oxygenation. In pulmonary haemorrhage, free haemoglobin in the airways could bind nitric oxide and thereby inhibit its effect. The anti-platelet effect of nitric oxide could be disadvantageous in acute haemorrhage.3 In a case of refractory hypoxia from diffuse lung haemorrhage caused by leptospirosis, oxygenation improved when nitric oxide was started on day 6 of intensive care admission,3 suggesting that residual blood in the airways does not contraindicate the use of nitric oxide.

A clinical trial of prone ventilation in ARDS6 found that oxygenation improved in 70% of instances, most improvement occurring in the first hour. The trial involved patients who fulfilled criteria for ARDS/acute lung injury to be placed prone for 6 h for 10 days. Overall mortality was not reduced, but patients with the lowest PaO2:FIO2 ratio had lower mortality at the end of the 10 day study, suggesting that ventilation in the prone position may be useful for severe hypoxaemia.6 The prone position may improve ventilation–perfusion (V/Q) matching by changes in inflation, ventilation, recruitment and perfusion of alveoli, both in normal patients and those with ARDS.79

Ventilation in the prone position may help conditions other than ARDS. Oxygenation improved in a patient with neurogenic pulmonary oedema when the patient was kept prone for 21 h without complication.10 The improved oxygenation often persists when the patient returns to the supine position, possibly because different pressures are required to open rather than keep open airways.8

Two case reports of acute large vessel pulmonary haemorrhage managed by placing the patient in the prone position emphasize the importance of the prone position in removing secretions and/or blood. These patients had haemorrhage from a main bronchus resulting in hypoxia with and without hypoventilation. In the supine position, secretions and blood will drain posteriorly blocking dependent airways and cause hypoventilation of dorsal lung units. Drainage of blood is encouraged by the prone position—the natural drainage position of the superior segments of the lower lobes.11 12 Clearing the airways to these lobes will allow greater ventilation and higher tidal volumes. Blood in the alveoli may be a barrier to gas diffusion. Drainage of this blood by the prone position may improve oxygenation.13

The patient reported here developed widespread alveolar haemorrhage from vasculitis, and prone ventilation improved oxygenation and ventilation dramatically. These responses were at least partially sustained in supine position. A review of the mechanisms for improvements in V/Q matching suggest why this may be so.69 Copious bloody secretions were noted each time the patient was placed prone, and the effectiveness of being turned prone to obtain drainage of the airways cannot be ignored.1113 There may be a tendency to ignore prone positioning for treatment after a study showing no mortality benefit in ARDS patients.6 This may be wrong. It may be inappropriate to extrapolate clinical findings from one condition—ARDS, to another—diffuse pulmonary haemorrhage. No studies exist of prone ventilation for pulmonary haemorrhage. The study of patients with ARDS may have failed to show benefit because the prone position was only used for short periods. Nitric oxide has been used to treat hypoxia from pulmonary haemorrhage,3 but was not considered for this patient. Other treatments for alveolar capillaritis such as factor VIIa may be considered in the future.5 Experience from this patient suggests that prone ventilation should be considered for severe hypoxia from diffuse pulmonary haemorrhage.


    References
 Top
 Abstract
 Introduction
 Case report
 Discussion
 References
 
1 Merkel PA, Choi HK, Niles JL. Evaluation and treatment of vasculitis in the critically ill patient. Crit Care Clin 2002; 18: 321–44[ISI][Medline]

2 Rodriguez W, Hanama N, Guy E, Guntupalli J. Pulmonary–renal syndromes in the intensive care unit. Crit Care Clin 2002; 18: 881–95[ISI][Medline]

3 Borer A, Metz I, Gilad J, et al. Massive pulmonary haemorrhage caused by leptospirosis successfully treated by nitric oxide inhalation and haemofiltration. J Infection 1999; 38: 42–5[CrossRef][ISI][Medline]

4 Hakanson E, Konstantinov IE, Fransson S-G, Svedjeholm R. Management of life threatening haemoptysis. Br J Anaesth 2002; 88: 291–5[Abstract/Free Full Text]

5 Bettensley AD, Yankaskas JR. Factor VIIa for alveolar haemorrhage in microscopic polyangiitis. Am J Resp Crit Care Med 2002; 166: 1291–2[CrossRef]

6 Gattinoni L, Tognini G, Pesenti A, et al. Effect of prone positioning on the survival of patients with acute respiratory failure. N Engl J Med 2001; 345: 568–73[Abstract/Free Full Text]

7 Pelosi P, Brazzi L, Gatinoni L. Prone position in acute respiratory distress syndrome. Eur Respir J 2002; 20: 1017–28[Abstract/Free Full Text]

8 Tobin A, Kelly W. Prone ventilation—it’s time. Anaesth Intensive Care 1999; 27: 194–201[ISI][Medline]

9 Mure M, Lindahl SGE. Prone position improves gas exchange—but how? Act Anaesthesiol Scand 2001; 45: 150–9[CrossRef][ISI]

10 Fletcher SJ, Atkinson JD. Use of prone ventilation in neurogenic pulmonary oedema. Br J Anaesth 2003; 90: 238–40[Abstract/Free Full Text]

11 Savage R. Prone position as a life saving measure for acute pulmonary haemorrhage in a young adult with cystic fibrosis. Anaesth Intensive Care 2002; 30: 223–5[ISI][Medline]

12 MacKenzie CF. Anatomy, physiology, and pathology of the prone position and postural drainage. Crit Care Med 2001; 29: 1084–5[CrossRef][ISI][Medline]

13 Savage R. Prone, head down for pulmonary haemorrhage. Br J Anaesth 2002; 89: 185–6[Free Full Text]