Incidence and Diagnostic Evaluation of Postoperative Fever in Pediatric Patients With Neuromuscular Disorders.

BACKGROUND: Children with neuromuscular disorders have a significant chance of developing scoliosis and/or hip dislocation during childhood and adolescence and often undergo surgical reconstruction. Because of their high rate of medical comorbidities, these patients are at higher risk of postoperative complications and are therefore often comanaged, along with orthopaedics, by pediatric medicine and critical care teams. Fever during the postoperative stay is a frequent occurrence and often prompts extensive diagnostic workup which increases the cost and burden on the health system but have unclear effects on the care of the patient. The objective of our study was to evaluate the frequency of postoperative fever in pediatric patients with neuromuscular disorders after spine and hip deformity correction surgery and evaluate the utility of the diagnostic tests performed for the assessment of postoperative fever.

METHODS: We performed an IRB-approved retrospective study on patients who underwent corrective surgery for neuromuscular scoliosis or hip dislocation at a single institution. The occurrence of postoperative fever was characterized by maximum temperature, postoperative day (POD) of occurrence, and frequency as defined by either single or multiple temperature spikes. Diagnostic tests performed for the workup of postoperative fever were reviewed. The cost per health effect was estimated by dividing the total costs of diagnostic tests by the number of tests that changed patient care.

RESULTS: In total, 108 patients (62 females and 46 males) were identified. A total of 42 patients (38.9%) underwent posterior spinal fusion and 66 (61.1%) patients underwent hip surgery (pelvic osteotomy in 8 patients, femoral osteotomy in 31 patients, and both pelvic and femoral osteotomy in 27 patients). The mean age at the time of surgery was 11.1 years (range, 3 to 18 y). In total, 66 patients (61.1%) developed postoperative fever (mean temperature, 38.6°C). The frequency of fever was in the form of multiple temperature spikes in 37 patients (56%) and in the form of a single spike in 29 patients (44%). Of the 149 diagnostic tests performed for postoperative fever, there were a total of 16 positive tests (10.7% of total tests ordered; n=16/149) including 5/27 urine analysis, 4/26 urine culture, 4/28 chest x-ray, 1/1 wound culture, 0/1 sputum culture, urine Gram stain 0/1, tracheal aspirate culture 0/1, throat culture 1/1, adenovirus polymerase chain reaction (PCR) 0/5, human metapneumovirus PCR 0/5, parainfluenza PCR 0/5, rhinovirus PCR 0/1, 1/3 bronchoalveolar culture, and 0/7 respiratory virus panel. A total of 37 blood cultures were drawn and all were negative. There was a significant difference (P=0.04) in frequency of negative diagnostic workup performed at the first, second, and third POD as compared with positive diagnostic workup. Total cost of the diagnostic tests was $65,284 and the cost per health effect was $6582 ($65,284/10). Diagnostic tests in patients with postoperative fever were associated with prolonged length of hospital stay in comparison with patients who did not perform any diagnostic workup using the Spearman ρ test (P=0.02).

CONCLUSION: In total, 61% of pediatric patients developed postoperative fever after surgical correction of neuromuscular deformity. An infectious source of postoperative fever was identified in 32.4% of patients with postoperative fever who underwent diagnostic workup. Urinary tract infection was the most common finding in patients with postoperative fever. Poor association between the development of postoperative fever and wound infection was found. Only 10.7% of fever diagnostic workup tests were positive in our population. The diagnostic workup tests might be less valuable if performed early on the first, second, and third PODs than those performed late after the third POD with exceptions based on clinical assessment. No patients with postoperative fever had positive blood cultures, therefore the routine use of blood cultures in the evaluation of postoperative fever in such population is not recommended.

LEVEL OF EVIDENCE: Level IV-retrospective.