| 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 Pediatrics and Pediatric Respiratory Medicine (Drs. Picard, Goldberg, Virgilis, and Schwartz), Infectious Disease Unit (Dr. Raveh), Shaare Zedek Medical Center, Jerusalem, affiliated to Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer Sheva; and Department of Pediatrics (Dr. Kerem), Hadassah Hebrew University Hospital, Mount Scopus, Jerusalem, Israel.
Correspondence to: Elie Picard, MD, Pediatric Pulmonology Unit, Shaare Zedek Medical Center, PO Box 3235, Jerusalem 91031, Israel; e-mail: Picard{at}szmc.org.il
Abstract
Study objective: To assess the effectiveness of one dose of dexamethasone (0.5 mg/kg; maximum, 10 mg) to prevent fever after bronchoscopy with BAL.
Design: Randomized, placebo-controlled study.
Patients: Immunocompetent nonfebrile children undergoing fiberoptic bronchoscopy with BAL.
Measurements and results: Sixty-nine children were included in the study. Thirty-eight children received saline solution, and 31 children received dexamethasone. The two groups were similar regarding the number of children < 2 years old, the percentage of abnormal bronchoscopic findings, the number of positive BAL culture findings, and the index of lipid-laden macrophages. Twenty-six children (68%) in the saline solution group (SG) had fever, compared to 3 children (9.6%) in the dexamethasone group (DG) [p < 0.001]. Fever after the procedure appeared later (12.3 ± 5.5 h) in the DG compared to 5.4 ± 2.7 h in the SG.
Conclusions: One dose of dexamethasone administered prior to performing bronchoscopy with BAL may prevent fever subsequent to the procedure. Further studies are necessary in order to determine the optimal dosing regimen for dexamethasone when used for this purpose.
Key Words: fever • fiberbronchoscopy in children • lavage • steroids
Fiberoptic bronchoscopy (FB) is a useful procedure to diagnose pulmonary diseases in childhood.1 The procedure is regarded as safe; however, rare serious complications such as arrhythmia, bleeding, transient hypoxemia, and bronchospasm have been reported.23 By contrast, transient fever in the hours following bronchoscopy is frequently noted, mainly when the procedure is performed with BAL. The fever, abrupt in onset, is usually accompanied by rigors and malaise, and responds to antipyretics.4 The incidence of postbronchoscopy fever varies according to different reports.5 We have reported in a prospective study6 an incidence of 48% in children undergoing FB with BAL. It has been shown that the transient fever is not due to bacteremia,7 although bacteremia and sepsis after FB with BAL may occur among immunodeficient patients.8910 Consequently, prophylactic antibiotics are not routinely recommended for the procedure. Several studies11 have demonstrated that the fever following bronchoscopy and BAL is related to the release of cytokines into the bloodstream by the alveolar cells. Prevention of this uncomfortable side effect, which usually warrants medical evaluation and laboratory investigations, and which may also alarm parents and caretakers, would be beneficial.
Steroids are known to decrease the release of cytokines from T-lymphocytes.12 The aim of our study was to examine prospectively the effect of one dose of corticosteroid administered prior to performing bronchoscopy and BAL on the incidence of fever following the procedure.
Materials and Methods
The study population included all children who underwent FB and BAL in the Shaare Zedek Medical Center from 2001 to 2003. Exclusion criteria were immunocompromised state, and concurrent treatment with steroids and fever > 38°C at the time of bronchoscopy. The study received the approval of the internal review board, and informed consent was obtained from the parents of each patient enrolled in the study. Children were randomized to receive either IV dexamethasone (0.5 mg/kg; maximum, 10 mg) or normal saline solution immediately prior to the procedure in a double-blind fashion. The solutions were prepared and administered to the patients on an alternating basis by the bronchoscopy nurse. The attending physicians were blinded to the group assignment. This dose of dexamethasone was chosen because it is similar to that used for prevention of postextubation airway obstruction.13 All procedures were performed with a bronchoscope (model BF-3C20, or BF-P200 for the older patients; Olympus; Tokyo, Japan). Sedation was achieved with midazolam (0.1 to 0.2 mg/kg) and propofol (1 to 3 mg/kg) administered IV. After local lubrication of the nostrils with lignocaine HCl 2% ointment, the bronchoscope was inserted through the nasal passages and advanced through the vocal cords into the tracheobronchial tree. Lidocaine 1% was atomized as needed for topical anesthesia through the channel of the bronchoscope at a maximum dose of 7 mg/kg. BAL was usually performed in the right lower lobe except in cases with localized pathology elsewhere. Three aliquots of 1 mL/kg of 0.9% sterile saline solution were instilled through the working channel after wedging the bronchus. Fluid was recovered by manual suction through a sterile specimen trap. All BAL fluid (BALF) specimens were tested in our microbiology laboratory for quantitative bacterial culture. In accordance with previous studies,14 we considered the BALF culture finding positive if there was growth of one bacterium at a concentration > 104 cfu/mL. BALF was also analyzed for lipid-laden macrophages (LLM) scoring to rule out pulmonary aspiration. For the scoring of LLM, the specimens were stained with oil red O, and an index as described by Colombo and Hallberg15 was assigned. An index > 90 was considered positive. After the procedure, the patients were monitored in the bronchoscopy unit by a nurse for a minimum of 4 h before discharge. An oral or rectal temperature during this time period > 38°C was considered fever.
Blood culture specimens were obtained in all children with a temperature > 38°C. All parents were instructed to monitor the body temperature during the 24 h following discharge by assessing for tactile fever which, if suspected, was to be confirmed by oral or rectal temperature measurement. Parents were advised to bring the child to the emergency department if fever occurred. After 24 h, one of the researchers contacted the family in order to ascertain that all episodes of fever were indeed reported, and to inquire if any other side effects were noted.
Statistical Analysis
We expected fever to develop in approximately 52.5% of patients undergoing BAL who received saline solution,6 and in approximately 15% of those who received dexamethasone. In order to achieve statistical significance at p < 0.05 and power 
90%, the calculated study size was 36 patients in each group. We applied the t test (nonpaired, two tailed) and 
2 test (Fisher Exact Test when applicable) for continuous and categorical variables as appropriate. Comparison of time (hours) until onset of fever was performed by the Wilcoxon signed-rank test. Significant p value was set at 0.05.
Results
During the study period, 178 children underwent FB in our unit and 69 of them were enrolled in our study. Reasons for exclusion were FB without BAL, fever at the time of FB, or refusal to participate in the study. All pertinent patient data are summarized in Tables 1, 2 . Thirty-eight children received saline solution, and 31 children received dexamethasone. The male/female ratios of the two groups were similar: 22 children (58%) in the saline solution group (SG) and 19 children (61%) in the dexamethasone group (DG) were male (p = 0.81). The average age (± SD) in the SG was 5.43 ± 5.3 years, slightly older than in the DG (3.1 ± 2.9 years) [p = 0.023]. The median age was similar in both groups: 3.28 years in the SG vs 2.22 in the DG (p > 0.05). The number of children < 2 years old did not differ between the two groups (SG, n = 12 vs DG, n = 13; p = 0.37). Thirteen patients in the SG group (34%) had a chronic disease, comparable to the 10 patients (47%) in the DG (p = 0.86) [Table 2]. Indications for FB in the SG were stridor and noisy breathing (n = 13), BAL for microbiology and pathology (n = 9), persistent pulmonary infiltrate (n = 8), persistent wheezing (n = 4), recurrent pneumonia (n = 3), and cyanotic spells (n = 1). Indications in the DG were stridor and noisy breathing (n = 12), BAL for microbiology and pathology (n = 9), persistent pulmonary infiltrate (n = 5), persistent wheezing (n = 4), and recurrent pneumonia (n = 1). Abnormal findings were observed in 28 patients (74%) in the SG and in 19 patients (61%) in the DG (p = 0.31) [Table 2]. There was significant bacterial growth in the BALF of 24 patients (68%) and 14 patients (45%) from the SG and DG, respectively (p = 0.152). In both groups, Haemophilus influenzae was the most frequent bacteria isolated (Table 2). LLM index was positive in 10 of 25 patients (40%) in the SG and in 7 of 25 patients (28%) in the DG (p = 0.55).
|
|
Discussion
In this double-blind, placebo-controlled study, one dose of dexamethasone prior to the procedure reduced the incidence of fever in children undergoing FB with BAL. The incidence of fever after FB with BAL in the placebo group was 68%, consistent with the 52% incidence we found in our previous study.6 This incidence was reduced to 9.6% when patients received one dose of dexamethasone.
Fever is a known complication of FB with BAL, occurring both in adults and in children. In adults, this complication has been reported with varied frequencies (2.5 to 16%).161718 Interestingly, a high incidence (65%) of fever after FB was noted in patients with nontuberculous mycobacteria in another adult study.19 Fever after FB has also been documented among healthy adult volunteers, with reported incidences of 21%20 and 50%.21 As in adults, the incidence of fever after FB with BAL in children varies according to different studies3622 (17 to 48%). Differences in the populations studied, methods of performing the FB and BAL, definition of fever, and data collection may explain the different incidences reported. Other important variables between the studies are the percentage of cases in which BAL was performed, and the degree to which the presence of fever was rigorously sought.
In our previous study,6 we identified three risk factors among children for fever after FB with BAL: abnormal bronchoscopic findings, positive BAL culture findings, and age < 2 years. In the study by Um et al,18 an abnormal bronchoscopic finding, prolonged procedure (> 10 min), severe bleeding, and a positive mycobacterial culture finding were more common among those with post-FB fever. Schellhase et al22 as well identified a positive BAL culture finding as a risk factor for high fever. In our present study, the two groups were similar for the presence of abnormal bronchoscopic findings and positive BAL culture findings. Although the average age of the SG was significantly older, the two groups were identical regarding the number of infants < 2 years old. Since none of the patients had severe bleeding or a positive culture finding for mycobacterium, these parameters could not be evaluated as confounding variables or as possible risk factors for post-FB fever.
The etiology of fever after FB with BAL is unclear. With the exception of rare cases of sepsis after FB with BAL among immunodeficient patients, infection does not seem to play a significant role.8910 Indeed, administration of prophylactic antibiotics before bronchoscopy did not affect the incidence of fever.23 Cytokines, mainly interleukin (IL)-1ß, IL-6, and tumor necrosis factor (TNF) are known to cause fever by raising the thermoregulatory set point in the hypothalamus.24 Studies have shown a rise in serum cytokines in febrile patients after bronchoscopy. Standiford et al11 found a significant elevation of TNF 4 h after bronchoscopy in a healthy volunteer with fever. Krause et al25 demonstrated a rise in IL-1ß, IL-6, and TNF in the serum of febrile patients 6 h following the procedure. Terashima et al26 assessed the WBC count and cytokine levels in healthy volunteers and among patients with pneumonia before and after FB with and without BAL. In both groups they found an increase in the peripheral WBC count following FB only when BAL was also performed. IL6 and granulocyte-colony stimulating factor levels were also increased after FB with BAL in both groups, but they were significantly higher in the patients with pneumonia compared to the normal volunteers. By contrast, there was no increase in IL-1, IL-8, or TNF levels in either group. These findings suggest that fever develops after bronchoscopy as a result of the release of pyrogenic mediators into the bloodstream due to an inflammatory reaction induced by BAL. This systemic inflammatory response is exaggerated in patients with underlying pulmonary disease. The source of the cytokines could be the macrophages located in the alveoli that are triggered during BAL by either the negative pressure applied, or by saline solution instillation.26
Glucocorticosteroids are known to suppress fever and IL-6 response due to an inflammatory state.27 Likewise, it was previously demonstrated that synthetic dexamethasone is a potent suppressor of intratumoral IL-6 production.28 Consequently, it is not surprising that we found a significant reduction in the incidence of fever in the dexamethasone group. Dexamethasone administered immediately prior to the bronchoscopy does not influence the BALF findings included BAL culture, but may indeed impede the release of cytokines by the activated macrophages that would otherwise occur a few hours after the procedure.
The mean time of onset of fever after FB in the SG was 5.4 ± 2.7 h, similar to the 4 to 9 h observed by De-Blic et al3 in their pediatric study. Interestingly, the fever in two of the three children in the DG was first noted 15 h after FB. Fever following FB with BAL typically develops a few hours after the procedure, and may persist for up to 24 h before resolving spontaneously.26 We speculate that the relatively late onset of fever in the two DG patients coincided with the waning effect of the dexamethasone. A second dose of dexamethasone administered 12 h after the procedure might fully eliminate post-FB fever. The third patient in the DG with fever had only a mild temperature elevation (38.2°C), which appeared earlier. It is possible that a higher dose of dexamethasone would be more effective and prevent all cases of post-FB fever.
A very high safety profile has been demonstrated for single-dose dexamethasone treatment in children with severe pain from pharyngitis,2930 as well as for viral croup.31 Moreover, short courses of dexamethasone are routinely administered in severe infections such as arthritis.32 Dexamethasone is also recommended for treatment of bacterial meningitis,33 with most studies34 showing no increase in adverse effects.
Antipyretics such as paracetamol might be suggested as being equally effective in preventing post-FB fever. However, this class of drug, which lacks antiinflammatory properties, could not be expected to prevent post-FB fever, which is due to an inflammatory process. Ibuprofen or other nonsteroidal antiinflammatory drugs might theoretically be a suitable alternative. Yet, since these medications are not commonly administered IV, and the patient must be fasting prior to the procedure, their use in this setting is precluded. Once nonsteroidal antiinflammatory drugs, such as ibuprofen, become available by the IV route, their use for this purpose should be considered.
In summary, we have demonstrated that fever after FB with BAL is a frequent occurrence in children. One dose of dexamethasone significantly reduced this untoward side effect. Further studies are necessary to determine the exact dosing of dexamethasone in order to achieve optimal results and to understand the mechanism by which it prevents the development of fever in children undergoing FB with BAL.
Footnotes
Abbreviations: BALF = BAL fluid; DG = dexamethasone group; FB = fiberoptic bronchoscopy; IL = interleukin; LLM = lipid-laden macrophages; SG = saline-solution group; TNF = tumor necrosis factor
None of the authors have any conflicts of interest to disclose.
Received for publication February 16, 2006. Accepted for publication July 7, 2006.
References
after bronchoscopy and bronchoalveolar lavage. Chest 1991;99,1529-1530This article has been cited by other articles:
|
|
 |
|
  P. Merkus, H. Janssens, M. Pijnenburg, H. Tiddens, and J. de Jongste Dexamethasone To Prevent Postbronchoscopy Fever in Children Chest, October 1, 2007; 132(4): 1413 - 1414. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
