The Efficacy of an Oscillating Bed in the Prevention of Lower Respiratory Tract Infection: DISCUSSION

We showed that continuous postural oscillation decreased the incidence of LRTI and pneumonia and shortened the median hospital stay for victims of nonpenetrating trauma. Patients in the control arm received standard nursing care, including turning from side to side every 2 h (although the efficacy of this effort to turn patients was sometimes compromised by the presence of skeletal traction devices). Although the mean ISS and APACHE-II scores were somewhat lower for the patients randomized to the study bed, the differences were small and not statistically signif-icant. It is unlikely, therefore, that differences between groups in severity of injury or physiologic derangement were responsible for the observed differences in incidence of LRTI and pneumonia. More patients with major head trauma were randomized to the control group, but it is unlikely that this can explain the observed effect of postural oscillation, since logistic regression analyses showed that treatment with a conventional hospital bed remained a significant risk factor for LRTI and pneumonia, even when severity of injury (ISS) and major head trauma were simultaneously considered in the model. Furthermore, when patients with and without major head trauma were analyzed separately, the conclusions were similar, although when LRTI was the outcome variable, the beneficial effect of the oscillating bed barely missed achieving statistical significance, possibly because of inadequate sample size.

Because it was impossible to perform this study in a blinded fashion, we established a priori a set of rigorous criteria for the diagnosis of pneumonia. As is customary, one criterion was the presence of infiltrates on the chest roentgenogram. Interpreting chest roent¬genograms in these critically ill blunt trauma patients however, was extremely difficult. Virtually all studies were obtained using portable equipment with the cassette posterior to the patient. Furthermore, pulmonary infection was only one of several possible etiologies for infiltrates on the chest roentgenogram; other possible causes included atelectasis, pulmonary contusion, adult respiratory distress syndrome and pulmonary edema due to fluid overload. In view of these difficulties, we also analyzed our data using a definition of pulmonary sepsis (LRTI) that relied on strict objective criteria not including interpretation of the chest roentgenogram. Using LRTI as the outcome variable, there was still a significant advantage for therapy with the RRKTT. We recognize that our definition of LRTI may have resulted in the inclusion of patients without bona fide pulmonary infections.

One weakness of the present study was the relatively large number of patients who were randomized to the study bed but were either never put on it or were taken off it prematurely. However, even when these 13 patients are excluded from the analysis of results, the main conclusions are the same: treatment with an oscillating bed significantly decreased the incidence of LRTI and pneumonia. Lack of familiarity with the bed during the early phase of the study was one reason for the relatively large number of patients who were randomized to the RRKTT but who were not placed on it. As we gained experience with the RRKTT, it became evident to us and our orthopedic colleagues that skeletal traction is effective on the oscillating bed. This view is supported by the literature. One patient with severe and uncontrollable intracranial hyperten¬sion secondary to head trauma was taken off the bed prematurely at the request of his family. Although data obtained by Gonzalez-Arias et al support the view that intracranial hypertension is not a contraindication for continuous postural oscillation, we were obviously unable to absolutely exclude the possibility that the study bed was contributing to elevated intracranial pressure in this particular patient. In several instances, patients initially tolerated the bed, but as they became more alert and physiologically stable, they found the bed confining and became agitated when the bed was rotated. Therefore, it was necessary to take these patients off the study bed prematurely. canadian pharmacy online

Our findings are consistent with results obtained in several previous related studies. Three studies are particularly convincing. Kelley et al, in a prospective randomized trial of continuous oscillation therapy for bedridden patients with acute stroke, studied 53 patients and found that the incidence of infection was decreased from 80 percent for patients managed in a conventional bed to 39 percent for patients managed using the RRKTT (p = 0.023). The incidence of pneu-monia was 52 percent and 28 percent for the control and study patients, respectively. Gentilello et al reported the results of a randomized, prospective trial of the RRKTT in 65 critically ill patients immobilized because of head injury and/or traction. In this study, the incidence of “major pulmonary complications” (ie, atelectasis plus pneumonia) was 65.8 percent in the control group and 33.3 percent in the RRKTT group (p<0.01). Pneumonia occurred in 34.2 percent and 18.5 percent of patients randomized to a conventional or oscillating bed, respectively (p value not reported). It is noteworthy that in the study by Gentilello et al, the incidence of pneumonia in the two groups was quite similar to the results obtained in the present study. Summer et al randomized 86 medical intensive care unit patients to either a conventional bed or the RRKTT. Although the overall mortality and incidence of pneumonia was similar in the two groups, certain subsets of patients (eg, those with pneumonia or chronic obstructive lung disease) benefitted from continuous postural oscillation, since in these sub-groups, ICU length of stay was significantly shortened.

Category: Disease

Tags: Oscillating Bed, Prevention, Tract Infection