Weaning from Mechanical Ventilation by Long-Term Nasal Positive Pressure Ventilation in Two Patients with Acute Respiratory Distress Syndrome Associated with Pneumococcal SepsisWindisch W. · Storre J.H. · Matthys H. · Sorichter S. · Virchow Jr. J.C.
Department of Pneumology, University Medical Clinic Freiburg, Germany Corresponding Author
Only few data concerning weaning by nasal positive pressure ventilation (NPPV) are available, and successful weaning by using NPPV in patients with acute respiratory distress syndrome (ARDS) and severe complications has not yet been described. Two cases with ARDS and both preexisting thoracopulmonary disease (infundibulum abnormality and suspected COPD) and associated complications (recurrent sepsis, acute renal failure, need for lobectomy, severe malnutrition) could not be weaned by invasive ventilatory techniques. Both patients presented with rapid shallow breathing and PaCO2 values >60 mm Hg during intermittent trials of spontaneous breathing, although the primary pathology and associated complications had been resolved. Patients were successfully adapted on NPPV in a stepwise approach after 93 days and 67 days of invasive ventilation. In one patient withdrawal from NPPV was possible after 2 months. In the other patient the duration of daily ventilation could be significantly reduced from 18 to 6 h/day after 9 months on NPPV. Therefore, patients with ARDS who cannot be weaned by invasive ventilatory strategies might be removed successfully from invasive mechanical ventilation by using NPPV even when there are preexisting thoracopulmonary disease and major complications during invasive ventilation.
Copyright © 2002 S. Karger AG, Basel
About 25% of patients admitted to the ICU with the need for intubation in order to manage acute respiratory failure are reported to have difficulties in weaning from mechanical ventilation after the underlying cause of acute respiratory failure has resolved [1, 2]. Most of them can be weaned by cost-intensive and time-consuming invasive weaning strategies. A minority, however, remains unweanable [1, 2]. These patients are believed to have a disruption of the balance of central respiratory drive, respiratory muscle capacity, and load of the respiratory muscle pump due to preexisting chronic thoracopulmonary disorders leading to hypercapnic respiratory failure (HRF) [2 , 3].
Nasal positive pressure ventilation (NPPV) has been increasingly used to treat chronic HRF and acute HRF , but also to assist weaning from mechanical ventilation [2, 5, 6, 7]. Most patients successfully weaned by NPPV have preexisting HRF due to chronic thoracopulmonary disorders such as COPD, chest wall deformities and neuromuscular diseases .
The acute respiratory distress syndrome (ARDS) is a common clinical syndrome. Its mortality is still high, although new methods of ventilatory support and new types of pharmacological support have been applied . After the acute phase the disease may resolve completely, but may also turn into fibrosing alveolitis with persistent abnormalities of pulmonary function . This may result in prolonged mechanical ventilation and weaning failure. However, successful noninvasive weaning strategies have not been described in patients with weaning failure due to ARDS with severe associated complications. Here, we present two cases with weaning failure after a prolonged period of invasive ventilation following ARDS due to pneumococcal pneumonia/sepsis who could successfully be weaned by NPPV.
The patients’ characteristics and initial clinical data are given in table 1. Both patients acquired pneumococcal pneumonia and developed septic shock with the need for administration of intravenous fluids and adrenergic agents controlled by pulmonary artery catheterization. Severe respiratory failure required early intubation and invasive ventilation (table 1). According to the susceptibility testing both patients were given penicillin. Patient A also received cefotiam, piperacillin/tazobactam, ciprofloxacin and fluconazole, since Staphylococcus aureus, Pseudomonas maltophilia and Candida were also isolated by bronchoalveolar lavage later in the course of her illness. Six weeks after the onset of the disease she developed a pseudopneumatocele with an abscess of the right lower lobe which was drained by thoracotomy, but a lobectomy of the right lower lobe was not performed. However, 1 week after the initial thoracotomy she underwent resection of the left lower lobe due to carnification. This was followed by a pulmonary leakage requiring positioning of a double-lumen endobronchial tube. In addition, she developed an acute renal failure with the need for hemodialysis for 1 week. In patient B the initial recovery was complicated by a second septic period due to infection with Enterobacter cloacae as assessed by bronchoalveolar lavage and urine catheterization. Initial susceptibility testing indicated piperacillin/tazobactam as treatment of choice, but this regimen had to be changed to ciprofloxacin due to no clinical improvement with temperatures up to 40°C.
Table 1. Patients’ characteristics, clinical data and ventilation parameters
Pneumococcal pneumonia/sepsis and associated complications eventually resolved sufficiently, but both patients continued to depend on mechanical ventilation. Tracheostomy was performed after 2 weeks. Although different types of weaning strategies including pressure-support ventilation, synchronized intermittent mandatory ventilation and intermittent trials of spontaneous breathing were used, the patients could not be weaned successfully. Patients developed HRF (PaCO2 >60 mm Hg) with rapid shallow breathing following periods of spontaneous breathing which required reconnection to the ventilator. In addition, stabilizing patients with invasive ventilatory techniques was complicated by the agitation of the patients with the need for sedative medication, by retention of secretion with the obstruction of cough and expectoration, and by hampered physiotherapy due to prolonged daily ventilation with shortened periods of spontaneous breathing.
Patients were therefore started on NPPV using a pressure-cycled ventilator (PV401, Breas®, Sweden). Before initiation of NPPV passive invasive ventilation aimed at maximal resting of the respiratory muscles was performed over a period of 24 h. Subsequently, patients were disconnected from the ventilator. A minimum of 1 h of spontaneous breathing was necessary in order to allow a safe establishing of NPPV. The tracheostoma was taped during periods of NPPV to allow nasal ventilation without major leakage. NPPV was used during the initiation with inspiratory pressures between 12 and 18 mbar and a low trigger threshold between –0.5 and –0.1 mbar (assisted ventilation). Supplemental oxygen was added to maintain SaO2 >95%. This was followed by a stepwise increment of the inspiratory pressure (20–30 mbar) until a further increase was not tolerated, and subsequently, the respiratory rate was increased beyond the spontaneous rate to establish passive NPPV while maintaining an I/E ratio of approximately 1/2 which was best tolerated by the patient. NPPV was most effective when the patients used an individual nasal mask. NPPV was initiated with short periods of 10–20 min at the beginning which were increased stepwise until invasive ventilation was completely replaced by NPPV.
After 5 days (patient A) and 7 days (patient B), respectively, patients were transferred from the ICU to a general ward in a clinically stable condition. Pulmonary function tests (Masterlab, Jaeger, Würzburg, Germany) and inspiratory mouth pressure measurements (ZAN®, Oberthulba, Germany) revealed severe restrictive pulmonary disease with impaired respiratory muscle function (table 2). Fibrosing alveolitis was established by computer tomography. NPPV was used for 14–18 h/day, most of which during the night. Patient A continued pressure-cycled ventilation with supplementation of 2 liters oxygen/min at home. The inspiratory pressure was 29 mbar, the respiratory rate was 21/min, and inspiratory time was 0.9 s on discharge. Arterial blood gas measurement (ABG) during NPPV revealed a PaCO2 of 41 mm Hg, a PaO2 of 78 mm Hg, an HCO3– of 28 mmol/l, and a pH of 7.47. Patient B was switched to volume-cycled ventilation because this was better tolerated (PV501, Breas, Sweden). She continued ventilation at home without any need for supplemental oxygen with a tidal volume of 0.75 liter, a respiratory rate of 26/min, and an I/E ratio of 1/1.5. ABG during NPPV revealed a PaCO2 of 41 mm Hg, a PaO2 of 83 mm Hg, an HCO3– of 26 mmol/l, and a pH of 7.41.
Table 2. Inspiratory mouth pressures, lung function parameters and BMI
Patient A discontinued NPPV after 2 months of noninvasive ventilation, since she remained normocapnic without further ventilation. ABG revealed a PaCO2 of 36 mm Hg, a PaO2 of 47 mm Hg, an HCO3– of 24 mmol/l, and a pH of 7.42 breathing room air; she, therefore, continued with long-term oxygen therapy with 2 liters oxygen/min at home. At 9 months of follow-up both patients had markedly improved inspiratory mouth pressures and lung function parameters indicating partial recovery (table 2). Patient A did not require long-term oxygen therapy any more, her ABG revealed a PaCO2 of 37 mm Hg, a PaO2 of 71 mm Hg, an HCO3– of 25 mmol/l, and a pH of 7.40 breathing room air. In patient B stepwise reduction of daily NPPV use was possible, but ventilation during the night (6 h) was necessary even after 9 months on NPPV due to hypercapnia breathing room air during the night with a PaCO2 of 50 mm Hg, a PaO2 of 79 mm Hg, an HCO3– of 29 mmol/l and a pH of 7.39.
Employing NPPV during weaning from mechanical ventilation has been shown to reduce weaning time, ICU stay, incidence of nosocomial pneumonia, and mortality in selected patients who required intubation in order to manage acute exacerbation of chronic HRF [2, 5, 7]. The mechanisms underlying this effect have been attributed to several factors. (1) Cough and expectoration are impaired by artificial airways predisposing to pulmonary infections, and prolonged invasive ventilation is associated with feeding aspiration. (2) In contrast to NPPV, patients being ventilated invasively are unable to speak. In addition, invasive ventilation techniques often require sedation and, thereby, prevent further communication. This might result in a prolongation of weaning. (3) An endotracheal tube markedly enhances the load on the respiratory pump which might lead to weaning failure in patients with preexisting impairment of the respiratory pump [3, 10, 11]. However, only few data concerning weaning by noninvasive ventilatory techniques are available. Furthermore, in selected patients with ARDS NPPV was associated with a successful outcome when established in the early phase , but successful weaning by using NPPV in patients with ARDS and associated complications has not yet been described.
In the present study two patients with ARDS due to pneumococcal sepsis and weaning failure were successfully adapted on NPPV in a stepwise approach. Both patients had HRF with inspiratory mouth pressures comparable to patients with acute HRF and those with weaning failure indicating an increased inspiratory load and decreased inspiratory muscle strength even after adequate treatment of pneumonia, sepsis and associated complications [13, 14, 15]. Decreased pulmonary compliance and increased alveolar dead space following fibrosing alveolitis might have contributed to HRF, but this was probably not the only condition which resulted in impairment of the respiratory pump, since fibrosing alveolitis due to ARDS, in general, leads to persistent hypoxemia, but not to hypercapnia . However, infundibulum abnormality of the chest, status postthoracotomy, and severe malnutrition probably led to further impairment of the respiratory pump. In addition, patient A was suggested to have COPD considering a history of heavy smoking (60 py), but preexisting lung function tests did not exist to verify this issue. According to the recovery phase of fibrosing alveolitis due to ARDS which is known to require several months  withdrawal from NPPV was possible in one case 2 months after onset of NPPV and was not possible in the second case even after 9 months on NPPV, although the duration of daily ventilation had been significantly reduced in that patient. Recovery was characterized by resolution of hypoxemia and hypercapnia following improvement of pulmonary function and inspiratory mouth pressures and additionally by an increase of the body mass index.
In conclusion, patients with ARDS who cannot be weaned by invasive ventilatory strategies might be removed successfully from invasive mechanical ventilation by using NPPV even when there are major complications during invasive weaning procedures. Therefore, NPPV should be considered as a treatment option for patients with weaning failure following ARDS. Prospective studies employing larger series are necessary to verify the benefit of NPPV in patients with ARDS and complicated weaning failure.
Wolfram Windisch, MD
Department of Pneumology, University Medical Clinic Freiburg
D–79106 Freiburg (Germany)
Tel. +49 761 270 3706, Fax +49 761 270 3704, E-Mail firstname.lastname@example.org
The authors disclose any and all financial involvement in any organization with a direct financial interest in the subject discussed in the submitted paper.
Received: Received: July 30, 2001
Accepted after revision: December 21, 2001
Number of Print Pages : 4
Number of Figures : 0, Number of Tables : 2, Number of References : 15
Respiration (International Review of Thoracic Diseases)
Founded 1944 as ‘Schweizerische Zeitschrift für Tuberkulose und Pneumonologie’ by E. Bachmann, M. Gilbert, F. Häberlin, W. Löffler, P. Steiner and E. Uehlinger, continued 1962–1967 as ‘Medicina Thoracalis’ as of 1968 as ‘Respiration’, H. Herzog (1962–1997)
Official Journal of the European Association for Bronchology and Interventional Pulmonology
Vol. 69, No. 5, Year 2002 (Cover Date: September-October 2002)
Journal Editor: C.T. Bolliger, Cape Town
ISSN: 0025–7931 (print), 1423–0356 (Online)
For additional information: http://www.karger.com/journals/res