| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Atelectasis is common after cardiac surgery and may result in impaired gas exchange. Continuous positive airway pressure (CPAP) is often used to prevent or treat postoperative atelectasis. We hypothesized that noninvasive pressure support ventilation (NIPSV) by increasing tidal volume could improve the evolution of atelectasis more than CPAP. One-hundred-fifty patients admitted to our surgical intensive care unit (SICU) with a Radiological Atelectasis Score 2 after cardiac surgery were randomly assigned to receive either CPAP or NIPSV four times a day for 30 min. Positive end-expiratory pressure was set at 5 cm H2O in both groups. In the NIPSV group, pressure support was set to provide a tidal volume of 810 mL/kg. At SICU discharge, we observed an improvement of the Radiological Atelectasis Score in 60% of the patients with NIPSV versus 40% of those receiving CPAP (P = 0.02). There was no difference in oxygenation (PaO2/fraction of inspired oxygen at SICU discharge: 280 ± 38 in the CPAP group versus 301 ± 40 in the NIPSV group), pulmonary function tests, or length of stay. Minor complications, such as gastric distensions, were similar in the two groups. NIPSV was superior to CPAP regarding the improvement of atelectasis based on radiological score but did not confer any additional clinical benefit, raising the question of its usefulness for altering outcome. IMPLICATIONS: This prospective, randomized, controlled study analyzing the treatment for atelectasis after cardiac surgery showed that noninvasive pressure support ventilation was superior to continuous positive airway pressure for improving atelectasis based on radiological score but did not confer any additional benefit, thus raising the question of its clinical usefulness.
The frequent incidence of atelectasis (54%92%) is a major concern after cardiac surgery because it contributes to the deterioration of pulmonary function and oxygenation (13). Multiple factors, such as pleural opening, postoperative diaphragmatic dysfunction, pain, immobilization, and bed rest, in addition to possible preexisting respiratory disease, are involved in the development of atelectasis in this clinical situation (35). Studies have compared different regimens of chest physiotherapy, incentive spirometry, or continuous positive airway pressure (CPAP) but have not separately analyzed prophylaxis from the treatment of postoperative atelectasis. Thus, the best treatment for atelectasis after cardiac surgery remains to be determined (2,6,7). CPAP is the most commonly used intervention, with a duration of treatment ranging from 25 breaths to 3 h continuously every day. Positive end-expiratory pressure (PEEP) levels applied range from 5 to 12 cm H2O (2,810). Noninvasive pressure support ventilation (NIPSV) is a recognized treatment for acute hypercapnic and hypoxemic acute respiratory failure (1113). There is some evidence that NIPSV improves oxygenation in postoperative nonhypercapnic patients, but NIPSV has not been studied for treating postoperative atelectasis (10,14,15). In proportion to the level of supplemental pressure support applied, it increases tidal volume, decreases respiratory rate, and improves gas exchange and diaphragmatic activity (16,17). Pressure support is usually set to achieve a tidal volume of 810 mL/kg (12,18). We hypothesized that NIPSV, by the adjunction of a pressure support to the usual CPAP, may be more effective for treating postoperative atelectasis after cardiac surgery. We tested this hypothesis in a prospective, randomized, single-blind, controlled study.
The protocol was approved by the institutional ethics committee, and written, informed consent was obtained from all patients. Between January 1999 and May 2000, consecutive patients admitted to our surgical intensive care unit (SICU) after cardiac surgery who had a Radiological Atelectasis Score 2 after tracheal extubation were included. The Radiological Atelectasis Score was defined according to Richter et al. (19): 0, clear lung field; 1, platelike atelectasis or slight infiltration; 2, partial atelectasis; 3, lobar atelectasis; and 4, bilateral lobar atelectasis. The exclusion criteria were pneumothorax, facial lesions, altered mental status, hemodynamic instability (defined as mean arterial blood pressure <60 mm Hg or a need for vasopressors, electrocardiograph instability with evidence of ischemia, or significant ventricular arrhythmia), planned secondary operation or transfer to another unit within 8 h after the diagnosis of atelectasis, more respiratory physiotherapy sessions than permitted in the study protocol requested by the clinician in charge or patients needing a permanent application of positive-pressure ventilation by mask to maintain a percutaneous oxygen saturation >90% (identified as refractory hypoxemia), or patient refusal.
Before the study inclusion, all patients admitted to the SICU after cardiac surgery were tracheally intubated and received synchronized intermittent mandatory ventilation with a tidal volume of 810 mL/kg body weight (respiratory rate, 12 breaths/min; inspiratory/expiratory time ratio, 1:2; PEEP, 5 cm H2O). Every ventilator was equipped with a heat and moisture exchanger. The respiratory rate and fraction of inspired oxygen (FIO2) were adjusted according to arterial blood gas analysis. As soon as the patient showed spontaneous breathing, the pressure support was reduced to 10 cm H2O, and the trachea was extubated according to the local protocol (percutaneous oxygen saturation >90% at an FIO2 of CPAP was provided by a gas mixer with an adjustable flow (Mov 60 O2/Air®; LNI Industry, Geneva, Switzerland), which was connected to a 5-L bag and a PEEP valve (Ambu, Copenhagen, Denmark) (20). The settings were flow rate 30 L/min, PEEP 5 cm H2O, and FIO2 adjusted to achieve a percutaneous oxygen saturation >90%. NIPSV was performed with Veolar FT® and Galileo® (Hamilton Inc., Reno, NV), Servo 300® (Siemens, Solna, Sweden), and Evita 4® (Dräger, Lübeck, Germany) in a spontaneous mode with a pressure support level adjusted to achieve a tidal volume of 810 mL/kg. Maximal pressure was delivered at 30 cm H2O, PEEP was 5 cm H2O, the minimal trigger flow was selected, and FIO2 was adjusted to achieve a percutaneous oxygen saturation >90%.
In both groups, patients underwent 30-min sessions four times a dayat 5:00 AM, 10:00 AM, 4:00 PM, and 10:00 PMin a semirecumbent position with the head at The preoperative values of arterial blood gas analyses, the vital capacity (VC) and the forced expiratory volume in 1 s (FEV1), age, sex, weight, and history of respiratory disease and smoking were recorded from the patients chart after randomization. The type of cardiac surgery and the duration of cardiopulmonary bypass were noted. Postoperative data included the simplified acute physiologic score version II (21), the duration of mechanical ventilation and the highest PEEP used during mechanical ventilation, the day of the first postoperative mobilization, the daily duration of the mobilization to an armchair, the day of thoracic tube removal, the SICU and hospital length of stay and mortality, and the readmission to the SICU.
The following assessments were performed at inclusion in the study (Time 0; T0) and every day (T1, T2, T3,...Tn) until the day of discharge (time of discharge; TD). Chest radiographs, performed in a semirecumbent position in bed or in the standing position whenever possible, were evaluated by two physicians certified in intensive care independent of the study and blinded to randomization. We assessed by Cohens
To achieve an expected improvement of the Radiological Atelectasis Score from 50% to 75%, the ß test with an
The flow diagram of patients in the study is summarized in Figure 1. Of 621 patients screened from January 1999 to May 2000, 225 (36%) met the radiological eligibility criteria. Seventy-five patients were excluded, and 150 were randomly assigned to receive CPAP or NIPSV. Fifteen patients discontinued the intervention: 9 patients in the CPAP group and 6 patients in the NIPSV group (see Fig. 1 for explanation). The reasons for discontinuation were comparable. Four patients in the CPAP group and three patients in the NIPSV group required more respiratory physiotherapy sessions than permitted in the study protocol for refractory hypoxemia. One patient in the CPAP group and no patient in the NIPSV group required tracheal intubation because of an acute respiratory failure. However, this was not statistically significant. The five pneumothoraces motivating the discontinuation of the study were due to an air leak during removal of the chest tube and not to the study intervention. However, all 150 patients were included in the intention-to-treat analysis.
At T0, the demographic and pre- and perioperative data of both groups of patients were similar (Table 1), and the two groups did not differ regarding pH, PaCO2, PaO2/FIO2 (Table 2), or the Radiological Atelectasis Score (Fig. 2, Table 3). The intrarater and the interrater variability of the Radiological Atelectasis Score were respectively 0.86 and 0.82 by Cohens test. At T0, an important decrease of VC and FEV1 occurred to a similar extent in both groups compared with preoperative values and persisted at TD (Table 2).
At TD, the Radiological Atelectasis Score was similar in both groups, but the differential Radiological Atelectasis Score ( Radiological Atelectasis Score = Radiological Atelectasis Score at T0 Radiological Atelectasis Score at TD) showed a statistically significant improvement in favor of the NIPSV group (P = 0.02) (Table 3). The Radiological Atelectasis Score decreased in 30 (40%) patients with CPAP, compared with 45 (60%) patients with NIPSV (P = 0.02). The Radiological Atelectasis Score even worsened in 8 (11%) patients with CPAP, compared with 1 (1%) patient with NIPSV (P = 0.03) (Fig. 2, Table 3). The difference in Radiological Atelectasis Score at TD comparing NIPSV with CPAP was 0.26 (95% confidence interval, 0.47 to 0.05), with a lower score among patients assigned to the NIPSV group (P = 0.01), adjusting for Radiological Atelectasis Score at T0. The Radiological Atelectasis Score at T0 was highly predictive of the score at discharge (coefficient, 0.6; 95% confidence interval, 0.390.82). At TD, pH, PaCO2, PaO2/FIO2, VC, and FEV1 were similar in the two groups (Table 2). Patients of the two groups received an equivalent number of treatment sessions (Table 4). The degree of comfort was similar in both groups (Table 4). No major complication, such as facial lesion due to the mask or pneumothorax induced by the intervention, was observed during the study. Minor complica-tions, such as radiological gastric distention or nau-sea, were equally distributed between the CPAP group (19 [25%]) and the NIPSV group (12 [16%]). No bronchoscopy was performed for treatment of atelectasis in any patient.
Other confounding factors potentially related to the efficacy of the treatment, such as the presence of hypersecretion, the number of induced cough sessions or nasotracheal suctionings, the cumulative fluid balance, thoracic drainage, mobilization, and pain intensity, were similar in the two groups (Table 4). The length of stay in the intensive care unit or in the hospital and the mortality rate during intensive care or in the hospital did not differ between the two groups (Table 2). No patient was readmitted to the SICU.
This large study investigated the treatment of atelectasis after cardiac surgery. We showed that NIPSVin-creases the rate of radiological resolution of atelectasis when compared with CPAP. However, this radiological improvement was not associated with any clinical amelioration, as shown by oxygenation and pulmonary function.
In the literature, studies on postoperative atelectasis after cardiac surgery do not clearly separate the effect of the prophylactic regimen to prevent atelectasis fromthe effect of the treatment of atelectasis. This difference is important in everyday clinical practice, and several authors have emphasized the need for studies analyzing separately the preventive and therapeutic regimens (8,23). This study is the first to analyze the therapeutic effect of NIPSV on atelectasis after cardiac surgery compared with CPAP. CPAP is often used for the prevention of postoperative atelectasis after cardiac surgery (2,10), but its efficacy in atelectasis treatment was evoked only after abdominal surgery: in 10 (83%) of 12 patients, Andersen et al. (9) showed a partial resolution of atelectasis after CPAP (PEEP, 15 cm H2O for Despite the radiological improvement of atelectasis, 149 patients left our SICU with persistent atelectasis. Furthermore, we could not find a difference between groups regarding oxygenation and pulmonary function that remained below normal limits in both groups at SICU discharge. This is consistent with the study of Ricksten et al. (26), who demonstrated that only the resolution and not the reduction of the radiological size of atelectasis can induce significant changes of oxygenation and pulmonary functions. NIPSV was as safe and comfortable as CPAP. These results on the safety and comfort of NIPSV are also consistent with other publications regarding NIPSV and acute respiratory failure (11,12). No major complication was noted, and few minor side effects were observed in both groups. The SICU or hospital length of stay and mortality did not differ between groups. However, our study was not designed or powered to address questionsabout the clinical outcome of the two therapeutic approaches. Our study has some limitations. The first regards the large number of patients who met exclusion criteria. Although this created a more homogeneous study group, some excluded patients, particularly those with refractory hypoxemia, could have benefited more from NIPSV. The second limitation was the short duration of the treatment of both groups because of the transfer of patients to the wards. Because of the restricted availability of SICU beds and for ethical reasons, the SICU stay was not to be prolonged because of the study. This could explain the persistence of radiological atelectasis in most of our patients, as in other studies (27). We cannot exclude that the pursuit of treatment by NIPSV after three days could have caused a further increase in resolution of atelectasis associated with an improvement of oxygenation or pulmonary function. Third, referring to previous studies, we chose only 5 cm H2O of PEEP (4,10), because our study sought to analyze the effect of the additional pressure support at the same level of PEEP in both groups. Therefore, our study cannot exclude that higher levels of PEEP may have improved the results in the CPAP or NIPSV group. Finally, the application of NIPSV is considered time consuming for nurses and may require more resources (12,28). This aspect of the comparison of the two methods was not addressed in our study. However, present preoccupations about cost effectiveness and costs of health care suggest that strong evidence of benefit for patients should be present whenever a new therapeutical approach is proposed. In conclusion, we showed in this large, prospective, randomized, single-blind, controlled study that NIPSV is superior to CPAP regarding the improvement of the Radiological Atelectasis Score after cardiac surgery. The persistence of atelectasis at SICU discharge and the lack of other clinical benefits (oxygenation, pulmonary function, and length of stay) raise the questions of the clinical usefulness and the duration of any treatment. Furthermore, NIPSV does not appear to be less resource consuming. Thus, we cannot encourage the use of NIPSV for the treatment of atelectasis after cardiac surgery.
The authors are grateful to the respiratory therapy team and the entire SICU team, who actively participated in this study.
Presented in part at the annual congress of the European Society of Intensive Care Medicine, Geneva, Switzerland, February 10, 2001.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|