The significance of post-operative fever following airway reconstruction

Results Forty-five percent of cases (35/78) had fever. Of those febrile, 46% ( n = 16) had significant fever. Overall, 20.5% had significant fevers. Fifty-two cases were single-stage LTP (SSLTP) with 31 febrile and 26 cases were double-stage LTP (DSLTP) with 4 febrile. SSLTP cases were at a significantly greater risk for post-operative fever compared with DSLTP, 59% vs 15% respectively ( p = 0.0002). 42% of febrile SSLTPs ( n = 13) had significant fevers compared to 50% ( n = 2) of febrile DSLTPs (Fisher's Exact p = 1.0). 81.5% of cases with CXR findings had fevers, but only 50% of these fevers were significant. Subjects with post-operative atelectasis were more likely to have a fever compared to subjects with no post-operative atelactasis (93% vs. 33% respectively, p < 0001). 30.8% of those with atelectasis had significant fever, compared to 52% of those without atelectasis ( p = 0.2) and 25 of SSLTPs vs. 3.9% of DSLTPs had atelactasis ( p = 0.027). No comorbidities were shown to be significant risk factors for post-operative fever. Conclusion Based on our review, most children undergoing LTPs will have insignificant fevers. Those children undergoing SSTLP and/or having post-operative atelectasis are at higher risk for post-operative fever. Fevers in children with double-stage procedures or all reconstruction cases with CXR findings other than atelectasis should have a thorough fever work-up. Keywords Airway reconstruction Fever Post-operative management 1 Introduction In the past 35 years, airway reconstruction has become the standard of care for moderate-to-severe subglottic stenosis in children. Laryngotracheoplasty (LTP) was first described in 1974 by Evans and Todd [1] , but was popularized by Cotton in the late 1970s as he introduced the use of costal cartilage grafting [2] . Laryngotracheal reconstruction has continued to evolve with the advent of the cricotracheal resection (CTR) and slide tracheoplasty procedures. As the operative management of subglottic stenosis has evolved, so has post-operative medical management. Laryngotracheal reconstruction can be accomplished as either a single-stage procedure, with the post-operative airway managed with an endotracheal tube, or as a double-stage procedure, in which the child's airway is managed with a tracheostomy. Management of children undergoing airway reconstruction has been well-described as there is a plethora of literature on the operative techniques and outcomes of the various procedures. Similarly, the post-operative pharmacologic management has also been well-described [3,4] . Fever is a common post-operative symptom following airway reconstruction. However, to our knowledge there is no literature regarding the management of fever in these patients. Usually, a fever in a pediatric intensive care unit elicits an extensive work-up, including chest radiographs and blood draws for complete blood counts and cultures to rule out sepsis. What is not known is whether these work-ups are necessary or whether they lead to a change in management in these airway cases. As a result, we conducted a retrospective review in an attempt to define the significance of post-operative fever following pediatric airway reconstruction. 2 Methods We conducted a retrospective analysis of 109 airway cases at Cincinnati Children's Hospital Medical Center from May 1, 2006–April 30, 2007. Thirty-one were excluded due to airway procedures other than LTP or charts were incomplete. Thus we had a 78 case cohort of LTP surgery examining post-operative fever management. The goal of this study was to determine the significance of post-operative fever following LTP. Fever was defined as temperature ≥38.5 and was determined “significant” if accompanied by a positive sputum, blood or urine culture. We also examined chest radiograph (CXR) results. CXRs were obtained in all children immediately post-operatively to rule out pneumothorax, but CXRs were subsequently obtained only if the patient had fever. Since this study was retrospective, the actual radiographs were not examined and we relied on the radiograph report for our findings analysis. Thus determination of atelectasis vs infiltrate or other radiographic findings were made by many different radiologists and thus there is an inherent potential for discrepancy. Severity of stenosis and prior airway surgery were examined. Comorbidities such as prematurity, seizure disorders, syndromes, developmental delay, cardiac anomalies, bronchopulmonary disease, gastroesophageal reflux disease and second airway disorder such as tracheomalacia or true vocal fold paresis also were analyzed. We did not examine the role of paralytics or sedation and their possible affect on post-operative fever. We examined both single-stage and double-stage reconstructions. Single-stage reconstruction patients typically were intubated between 5 and 7 days, while double-stage reconstructions relied on tracheostomy airway with suprastomal stenting. All patients had neck and chest drains placed at the time of surgery with both drains typically removed by post-operative day 3. All patients had prophylactic antibiotic coverage until the drains were removed. 3 Results The mean age of the patients was 63.5 months (range 5-264), with 19% of patients have grade II stenosis, 74% with grade III and 7% with grade IV stenosis. Forty-five percent of cases (35/78) had fever. Of those febrile, 46% ( n = 16) had significant fever, thus 16/78 (20.5%) of patients had significant fevers. ( Fig. 1 ) Significant fevers were due to positive sputum cultures in 81% (13/16), positive blood cultures in 25% ( n = 4), and positive urine culture in 6% ( n = 1) of cases. Two patients had both positive sputum and blood cultures. Only 4 patients had an elevated white blood cell count and only one of these was associated with a positive culture result. Neither patient age, stenosis grade nor prior airway surgery were determined to be significant regarding fevers. Fifty-two cases were single-stage LTP (SSLTP) with 31 of these having fever. Twenty-six cases were double-stage LTP (DSLTP) with 4 having fever. SSLTP cases were at a significantly greater risk for post-operative fever compared with DSLTP, 59% vs 15% respectively ( p = 0.0002). Twenty-nine percent of febrile SSLTPs ( n = 9) had significant fevers compared to 50% ( n = 2) of febrile DSLTPs (Fisher's Exact p = 1.0). ( Fig. 2 ) Of those with fever, 81% had positive chest radiograph findings (such as atelectasis, pneumonia, “consolidation,” “opacification,” or pleural effusion), but only 50% of these fevers were significant. Atelectasis was the most common post-operative CXR finding, however only 31% of those with atelectasis had significant fever, compared to 52% of those with CXR results other than atelectasis ( p = 0.2). SSLTPs were much more likely to have atelectasis compared with double-stage cases, 25% vs 3.9% ( p = 0.027), respectively. No comorbidities were shown to be significant risk factors for post-operative fever. 4 Discussion Pediatric airway reconstruction post-operative management can be a tricky and often frustrating experience. There is no consensus on how to manage these patients. Decisions regarding antibiotic use, when to pull drains, use of sedation, when to extubate or remove stents are based mostly on experience and in most cases individualized to each reconstruction. Anecdotally, it seemed that almost all airway reconstruction patients develop a fever at some point in their first week post-operatively, and to our knowledge there is no literature examining this type of fever management. The argument over what constitutes a significant fever could lead to endless debate, but we elected to define a significant fever as one that had a positive culture result, and thus would warrant a change in management. This study was borne from the experience that most patients with laryngotracheoplasties develop post-operative fever and, as a result, undergo a fairly expensive fever work-up. At Cincinnati Children's Hospital the cost of a fever work-up, which includes sputum, blood and urine culture as well as CXR is $1300. At Cincinnati Children's Hospital, post-operative airway surgery management is fairly uniform. SSLTP patients are typically extubated about a week after surgery. Antibiotic coverage is usually discontinued once the chest and neck drains are removed, which is usually by post-operative day 3. In cases where the patient had a prior history of MRSA, antibiotic coverage is usually continued in the form of Bactrim, Clindamycin or Vancomycin. Positive chest radiograph findings such as atelectasis, pneumonia, pleural effusion, consolidation or infiltrate can be concerning findings, but unless accompanied by a positive sputum culture, do not warrant increased antibiotic coverage. Based on our review, about half of all post-operative LTP patients have post-operative fever and of those, about half have significant fever. While most children undergoing LTPs will have insignificant fevers, those children undergoing SSTLP appear to be at higher risk for post-operative fever. However, only 29% of febrile SSLTPs ( n = 9) had significant fevers compared to 50% ( n = 2) of febrile DSLTPs (Fisher's Exact p = 1.0, p = 0.0002). We also found that the majority of significant fevers had positive sputum results (81%), with much fewer having positive blood (25%) or urine (6%) cultures. While 81% of post-operative LTP patients with fever had positive radiographic findings, only half of these fevers were significant. Atelectasis was the most common post-operative CXR finding, however only 31% of those with atelectasis had significant fever, compared to 52% of those who had other radiographic findings (such as pneumonia, infiltrate, etc.) ( p = 0.2). We found that SSLTPs were much more likely to have atelectasis, but the true number of patients with atelectasis may be higher in both types of LTPs since CXRs were only obtained if fever was present. Comorbidities such as prematurity, seizure disorders, syndromes, developmental delay, cardiac anomalies, bronchopulmonary disease, gastroesophageal reflux disease and second airway disorder such as tracheomalacia or true vocal fold paresis due not appear to be significant risk factors for post-operative fever in this review. There have been several studies [3,4] that have demonstrated an increase in pulmonary complications with the use of neuromuscular blockade and sedation. However, the use of paralytics and sedation has long been thought to be necessary in single-stage procedures to prevent unplanned extubations, to allow post-operative swelling to subside and to help stent the graft while it undergoes epithelialization. Jacobs et al. demonstrated that SSLTP patients who did not require physical restraints, paralytics or sedation had significantly less atelectasis compared with those children who were kept immobile and sedated post-operatively. In addition, when atelectasis was present, it improved significantly quicker in the non-sedated group [4] . Bauman et al. found atelectasis to be more common with patients with a history of bronchopulmonary dysplasia. They also found that those patients undergoing intermittent paralysis, as opposed to continuous paralysis, had less atelectasis and a shorter time to extubation. Unfortunately, our review was unable to accurately ascertain which of the single-stage procedures received post-operative sedation and to what extent sedation was used. Some patients were completely sedated, while others appeared to be sedated for part or none of the post-operative period. Intuitively, the SSLTP patients tend to receive more sedation (sometimes complete paralysis) compared with the DSLTP patients, who require no post-operative sedation. Given the fact that pulmonary toilet and mucociliary clearance is much easier in a trach patient compared with an intubated patient, it is not surprising that more than half of our SSLTPs had fever and were at a significantly greater risk for post-operative fever compared with DSLTP, 59% vs 15% respectively. SSLTP also had a significantly greater risk for atelectasis compared with double-stage cases, 25% vs 3.9% ( p = 0.027), respectively. Despite this, our atelectasis rate is low compared with literature reports up to 60% [5] , but this maybe due to fewer radiographs since only those with fever had CXRs. 5 Conclusion Our recommendations are that fevers in children following double-stage procedures or all reconstruction cases with CXR findings other than atelectasis should have a thorough fever work-up. Single-staged patients should be managed with minimal to no sedation or paralytics if possible since it has been shown in numerous studies to lead to increase post-operative complications. We also recommend further study in the form of a randomized trial to clarify the significance of post-operative fever in airway reconstruction and to determine if treatment is necessary. References [1] J.N.G. Evans G.B. Todd Laryngotracheoplasty J. Laryngol. Otol. 87 1974 589 597 [2] R.T. Cotton Management of sublgottic stenosis in infancy and childhood. Review of a consecutive series of cases managed by surgical reconstruction Ann. Otol. Rhinol. Laryngol. 87 1978 649 657 [3] B.R. Jacobs B.A. Salman R.T. Cotton K. Lyons R.J. Brilli Postoperative management of children after single-stage laryngotracheal reconstruction Crit. Care Med. 29 2001 164 168 [4] N.M. Bauman T.L. Oyos D.J. Murray S.C. Kao M.J. Biayati R.J. Smith Postoperative care following single-stage laryngotracheoplasty Ann. Otol. Rhinol. Laryngol. 105 1996 317 322 [5] K. Stenson R. Berkowitz T. McDonald B. Gruber Experience with one-stage laryngotracheal reconstruction Int. J. Pediatr. Otorhinolaryngol. 27 1993 55 64