| 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 Department of Clinical Anesthesiology (Drs. Kindgen-Milles, Buhl, Gabriel, and Müller); and the Department of Vascular Surgery and Kidney Transplantation (Dr. Böhner), Heinrich-Heine-University Düsseldorf, Düsseldorf, Germany.
Correspondence to: Detlef Kindgen-Milles, MD, Department of Clinical Anesthesiology, Heinrich-Heine-University Düsseldorf, Moorenstr. 5, D-40001 Düsseldorf, Germany; e-mail: kindgen{at}uni-duesseldorf.de
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Design: Prospective interventional study.
Setting: Surgical ICU of a university hospital.
Patients: Twenty consecutive patients after thoracic, abdominal, or combined thoracoabdominal operations, in whom pulmonary oxygen transfer deteriorated continuously following elective extubation after initial mechanical ventilation. Respiratory failure was due to atelectasis and/or left heart failure, and all patients met predefined criteria for reintubation.
Interventions: nCPAP therapy (8 to 10 cm H2O) was initiated if PaO2 had decreased to < 80 mm Hg despite application of 100% oxygen (flow, 25 L/min), intermittent mask continuous positive airway pressure, and maximum conventional therapy.
Measurements and results: nCPAP treatment was started 24.1 ± 3.4 h after elective extubation. PaO2 was < 80 mm Hg in all patients, in 13 patients it was < 60 mm Hg, and in 3 patients it was < 50 mm Hg. Mean PaO2/fraction of inspired oxygen (FIO2) ratio had decreased to 60 ± 2.6, and increased within the first hour of nCPAP to 136 ± 12 (p < 0.001). The clinical condition in all patients improved further, and after 35.2 ± 6.3 h, all patients were well oxygenated by face mask at ambient pressure (PaO2/FIO2 ratio, 146 ± 14). Two patients were reintubated for reasons unrelated to oxygenation or ventilation (data are presented as mean ± SEM).
Conclusions: nCPAP is safe, easy to apply, and effective to improve arterial blood oxygenation in < 1 h in postoperative patients with severe nonhypercapnic oxygenation failure. In these patients, who otherwise would have been reintubated, nCPAP can avoid endotracheal reintubation and mechanical ventilation.
Key Words: acute respiratory failure • continuous positive airway pressure • endotracheal intubation • hypoxemia • noninvasive ventilatory support
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Avoiding endotracheal intubation (EI) and mechanical ventilation (MV) by noninvasive ventilatory support can significantly reduce infectious complications,8 duration of ICU therapy,8 9 and mortality9 in patients with acute respiratory failure of various causes, so that the benefits of avoiding EI and MV in patients with acute respiratory failure are generally accepted now. To avoid reintubation in patients after elective extubation following MV is even more important, because reintubation in this patient group not only prolongs ICU and hospital stay, but most importantly, increases mortality rate more than sevenfold.10
In three randomized clinical trials, mCPAP improved the clinical condition and reduced intubation rate in patients with pulmonary edema.3 11 12 However, in these studies, the need for immediate EI and MV at times seemed questionable, and only 35 to 65% of all patients in the control groups finally met intubation criteria.3 11 12 We employed nCPAP as a strategy to avoid reintubation and MV in postoperative patients who, at study entry, all met predefined criteria for reintubation because of severe hypoxemic respiratory failure following initial MV and elective extubation. We show that with the exceedingly simple technique of nCPAP, reintubation was avoided.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
5 cm H2O;
normocapnia, with pressure support 5 cm H2O;
rectal temperature > 36.6°C, and a stable cardiovascular system
(ie, systolic BP > 100 mm Hg, heart rate < 120
beats/min, central venous pressure 6 to 12 mm Hg, no metabolic
acidosis, spontaneous diuresis > 1 mL/kg/h and no increase in
catecholamine dose in the last 3 h). Following extubation, standard treatment included regular chest physiotherapy, body repositioning, and application of intermittent positive airway pressure by face mask for 10 min every 2 to 4 h. N-acetylcysteine, 900 mg/d, and ambroxol, 45 mg/d,13 were applied IV. Inhalation of fenoterol and ipratroprium bromide was performed every 4 h.
Patients who developed cardiac failure (rales on chest auscultation, decrease in urine output < 1 mL/kg/h without hypovolemia, signs of pulmonary congestion in chest radiograph, pulmonary edema) were treated with IV dobutamine, nitroglycerin, and furosemide, as needed. Oxygen was supplemented via a nonocclusive face mask (flow, 25 L/min, heated and humidified) sufficient to achieve a PaO2 > 70 mm Hg. Criteria for reintubation due to hypoxemia were as follows: PaO2 < 80 mm Hg while on 100% oxygen (flow, 25 L/min), or PaO2/FIO2 ratio < 80, plus either a respiratory rate > 30 breaths/min, or a progressive (> 3 h) decline in oxygenation documented by blood gas analyses.
In those patients who met reintubation criteria, were conscious, and
were able to protect their airways, we started nCPAP. The same source
of oxygen/air that was used to apply oxygen at ambient pressure was now
used to provide a constant flow (60 L/min, heated and humidified),
passing through a compliant rubber bag (volume, 10 L) via a T piece to
the nasal mask (Respironics; Pittsburgh, PA). End-expiratory pressure
was measured at the T piece, and kept constant at 8 to 10 cm
H2O by means of a standard positive
end-expiratory pressure valve. During the first day, nCPAP therapy was
never interrupted for > 30 min. Later, for nursing care, we allowed
three to four periods of 30 min without nCPAP.
Failure of nCPAP therapy, which according to the protocol would lead to EI and MV, was defined as follows: intolerance or refusal of therapy by the patient (not if meanwhile oxygenation had improved sufficiently), lack of increase of PaO2 > 50 mm Hg within the first hour or a further decrease of PaO2 despite nCPAP, or the need to increase catecholamine therapy within the first hour.
Exclusion criteria were as follows: refusal to participate, lack of consciousness, inability to protect the airways, age < 18 years, pregnancy, deformities of the upper airways, acute bleeding of the upper airways, severe pulmonary emphysema with bullae, hypercapnic respiratory failure, severe sepsis, and hypoxemic respiratory failure after accidental self-extubation.
All patients had arterial and central venous catheters in place. Arterial blood gases were measured before and after 1 h of nCPAP, and thereafter at least every 3 to 6 h. After the first hour, FIO2 was adjusted to keep a PaO2 > 70 mm Hg. nCPAP was discontinued if patients were oxygenated sufficiently during a trial of a nonocclusive face mask at ambient pressure (PaO2 > 70 mm Hg; FIO2 < 0.7; flow, 25 L/min/). Hemodynamic variables (ie, arterial BP, central venous pressure, heart rate) were recorded continuously.
From all arterial blood gas analyses, PaO2/FIO2 ratios were calculated for each patient before, during the first 24 h, and after cessation of nCPAP, with the FIO2 taken as that chosen at the oxygen blender, which had previously been calibrated.
The study was performed in accordance with the declaration of Helsinki. The study protocol was approved by the Committee on Medical Ethics of the Heinrich-Heine-University, and informed consent was obtained from the patient or his or her relatives. Data before, during, and after cessation of nCPAP were compared, with each patient serving as his own control. For statistical testing, analysis of variance for repeated measurements with Scheffe’s post hoc comparison, and the two-tailed t test with Bonferroni’s correction for multiple comparisons were used. Significance was accepted as p < 0.05. Data are presented as mean ± SEM.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
nCPAP was initiated 24.1 ± 3.4 h after extubation. Mean PaO2 before nCPAP-therapy was 60 ± 2.6 mm Hg; in 13 patients, PaO2 was < 60 mm Hg; in 3 patients, it was < 50 mm Hg. nCPAP improved arterial blood oxygenation within the first hour, more than doubling mean PaO2 to 132 ± 17.6 mm Hg (p < 0.0001; Fig 1 , left, A), even though in seven patients, FIO2 had been lowered (in violation of the protocol) already within the first hour (mean FIO2 0.85 ± 0.05; p < 0.01; Fig 1 , center, B). In two patients, PaO2 increased only slightly in the first hour (from 48 to 53 mm Hg and from 45 to 59 mm Hg), but the clinical condition of these patients, both of whom were receiving dobutamine, improved; 3 h later, the PaO2 was > 60 mm Hg. In all but two patients, respiratory rate decreased (p < 0.004; Fig 1 , right, C).
|
|
Two patients were reintubated, none of them for reasons related to hypoxemia or hypercapnia. One female patient (thoracoabdominal aortic aneurysm) developed acute renal and hepatic failure, became somnolent, and was reintubated to protect her airways. In a second patient, there was a leak from surgical anastomosis after gastrectomy for cancer. She was reintubated for operative revision. Both patients died later of multiorgan failure following septic shock. At the time of EI, while on nCPAP, in both of these patients oxygenation and ventilation were adequate. A third patient had been discharged to the general ward after resolution of his respiratory failure, but returned 2 weeks later with a deep sternal wound infection. He also died in septic multiorgan failure weeks later.
In general, nCPAP was well tolerated. No patient complained of distress or refused treatment. There was no evidence of gastric distension, nausea or vomiting, or conjunctivitis. In four patients, we found ulcerations of the dorsum nasae, which were severe in three patients. In all three, the nasal mask had been in place without interruption for > 24 h.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Following extended cardiac, vascular, or abdominal surgery, patients may develop pulmonary complications, such as atelectasis,14 consolidation of lung areas, or pneumonia.15 Pulmonary gas exchange can additionally be impaired by left heart failure following extracorporeal circulation or high cross-clamping of the aorta.16 In a clinical setting, it is sometimes difficult to decide which of these impairments contributes more to the deteriorating condition of an individual patient. Therefore, and because all patients with prevailing cardiac failure also had some degree of atelectasis, we chose not to separate our patients into primary cardiac or pulmonary failure. A combined pulmonary and cardiac disturbance may initiate a vicious cycle to result in profound decrease in FRC followed by hypoxemia, and an increase in the WOB leading to respiratory muscle fatigue.17 18 Supplemental oxygen is commonly used in high-risk patients and may prevent arterial hypoxemia; however, it is a symptomatic but not a therapeutic approach. Instead, treatment should be directed to increase lung volume, to reduce WOB, and to improve cardiac performance, all of which—for the following reasons—can be accomplished effectively by increasing intrathoracic pressure.
Increasing intrathoracic pressure can restore a decreased FRC and thus improve pulmonary oxygen transfer, both by means of mCPAP,2 and nCPAP.2 However, Lindner et al5 showed in patients after laparotomy that increases in FRC following 3 h of mCPAP are not sustained to the next day. Stock et al4 demonstrated that after sternotomy, mCPAP-induced increases of FRC do not sustain > 10 min following cessation of treatment. This explains why intermittent mCPAP could not stop or reverse developing hypoxemia in our patients, and continuous forms of therapy are necessary.
CPAP may reduce the WOB by shifting the pressure-volume curve to the right, which is particularly important in patients with abdominal distension after laparotomy. In patients with cardiac failure, Lenique et al6 showed that mCPAP with 5 cm H2O and 10 cm H2O significantly reduced WOB. clinical sign of a reduced respiratory workload is a decrease of respiratory rate, which was observed in 90% of our patients. Although CPAP is not a noninvasive method of ventilation, it thus can unload respiratory muscles and avoid respiratory muscle fatigue.19
CPAP may improve cardiac performance in acute or chronic congestive heart failure by three mechanisms. In patients with limited cardiac reserve secondary to coronary artery disease, increased arterial oxygen content also increases myocardial oxygen supply. By decreasing venous return, CPAP can reduce preload. Probably more important is that elevated intrathoracic pressure decreases negative intrathoracic pressure swings during inspiration.7 This way, left ventricular transmural pressure and afterload are reduced, and cardiac index can be augmented via increased stroke volume.7 20 21 Accordingly, we observed substantial clinical improvement in two patients in spite of only minor increases in blood oxygenation during the first hour of nCPAP. Since these observations cannot be attributed to increased arterial oxygen tensions, they most likely result from an improvement in left heart function, even more so since both patients were receiving dobutamine.
From the above, we can conclude that increasing intrathoracic pressure by noninvasive means is a reasonable therapeutic approach to support patients with combined pulmonary and cardiac failure. In contrast to patients with hypercapnic respiratory failure, in those with nonhypercapnic hypoxemia, nCPAP alone may improve clinical condition and blood oxygenation, so that the more-sophisticated techniques of noninvasive ventilation, which require a ventilator, may not be necessary. This approach is supported also by clinical studies. Dehaven et al1 had treated hypoxemia following general surgery with mCPAP. While arterial blood oxygenation improved, these patients had an FIO2 of 0.45, a PaO2/FIO2 ratio > 130 at study entry, and EI was not indicated.1 More recently, in studies by Bersten et al3 and Lin et al,11 in patients with pulmonary edema that were treated with mCPAP, the intubation rate was lower compared to those receiving oxygen at ambient pressure. Again, compared to our study group, those patients had less-severe hypoxemia, and only 35 to 50% of the patients who received oxygen alone met their less-restrictive intubation criteria.3 11 Thus, the present study is unique with respect to the fact that all patients met generally accepted intubation criteria already at study entry. Avoidance of reintubation in patients following elective extubation is of great importance, because mortality increases more than sevenfold with renewed EI and MV.10 According to our results, nCPAP in addition to extensive standard therapy can obviate the need for reintubation in such patients.
Common criticism of studies as the one presented here concern the lack of a randomized control group, and our conclusions rest on the premise that patients were candidates for reintubation without nCPAP-therapy. Admittedly, we must accept that physicians differ in the criteria they use for deciding whether a patient requires EI or not. The important point is that regardless of the actual PaO2, blood oxygenation had deteriorated continuously over hours in spite of extended standard therapy. One must conclude that only a few physicians would postpone EI in such patients, waiting for further arterial desaturation to perform intubation in severe hypoxemia. After starting nCPAP, improvement in blood oxygenation usually occurs rapidly, ie, within minutes up to 1 h,3 11 12 which was also observed in our study. Institution of nCPAP with a continuous flow system is simple and can be started within minutes, not only in the ICU, but also in intermediate care wards. Thus, an effective means to increase intrathoracic pressure is rapidly available. A trial of nCPAP can hence be started in almost any patients with hypoxemic respiratory failure before performing EI, especially in the absence of severe hypercapnia and respiratory acidosis. It probably should be undertaken in patients with deteriorating pulmonary oxygen transfer before hypoxemia is manifest, unless immediate EI is required for other causes. It should be kept in mind, however, that if pulmonary oxygen transfer does not improve within the first hour, further improvements are unlikely and EI and MV should not be delayed any further.
The rationale to use nCPAP and not mCPAP rests on clinical studies that show both techniques to be equally effective to increase FRC.2 In contrast to mCPAP, nCPAP does not hinder verbal communication, causes less claustrophobia, is better tolerated by patients, and rated significantly more comfortable.3 20 21 22 To achieve a real continuous therapy without interruption, we therefore chose to use nCPAP. Finally, in-hospital mortality in this patient group was lower than predicted. However, it remains to be shown in larger study groups whether the avoidance of EI and MV in nonhypercapnic respiratory failure by nCPAP reduces morbidity and mortality to the same extent as noninvasive ventilation techniques undoubtedly did in hypercapnic respiratory failure.
In summary, nCPAP substantially increased systemic blood oxygenation within 1 h in patients with severe nonhypercapnic oxygenation failure following elective extubation after cardiac, vascular, and abdominal surgery. We showed that this exceedingly simple technique is safe and effective to avoid EI and MV, because at study entry, all patients met predefined criteria of intubation. Since nCPAP can be initiated also in intermediate care wards, it may reduce readmission rate to the ICU, and thus resolve some capacity problems of ICUs.
|
Footnotes |
|---|
Received for publication March 16, 1999. Accepted for publication October 15, 1999.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Jaber, J.-M. Delay, G. Chanques, M. Sebbane, E. Jacquet, B. Souche, P.-F. Perrigault, and J.-J. Eledjam Outcomes of Patients With Acute Respiratory Failure After Abdominal Surgery Treated With Noninvasive Positive Pressure Ventilation Chest, October 1, 2005; 128(4): 2688 - 2695. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Kindgen-Milles, E. Muller, R. Buhl, H. Bohner, D. Ritter, W. Sandmann, and J. Tarnow Nasal-Continuous Positive Airway Pressure Reduces Pulmonary Morbidity and Length of Hospital Stay Following Thoracoabdominal Aortic Surgery Chest, August 1, 2005; 128(2): 821 - 828. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Takami and H. Ina Beneficial effects of bilevel positive airway pressure after surgery under cardiopulmonary bypass Interactive CardioVascular and Thoracic Surgery, June 1, 2003; 2(2): 156 - 159. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. McGrath, M. Das-Gupta, and G. Clarke Bronchoscopy via Continuous Positive Airway Pressure for Patients With Respiratory Failure Chest, February 1, 2001; 119(2): 670 - 671. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
