C-reactive Protein as a Potential Biomarker of Residual Obstructive Sleep Apnea Following Adenotonsillectomy in Children.

STUDY OBJECTIVES

Adenotonsillectomy (AT) is first-line treatment for pediatric obstructive sleep apnea (OSA), with most children having improvements in polysomnography (PSG). However, many children have residual OSA following AT as determined through PSG. Identification of a biomarker of residual disease would be clinically meaningful to detect children at risk. We hypothesize serum high-sensitivity C-reactive protein (hsCRP), an inflammatory biomarker, is predictive of residual OSA following AT.

METHODS

PSG was performed both preoperatively and postoperatively on children undergoing AT for the diagnosis of OSA. HsCRP serum concentrations were determined in all children pre-AT, and in most children post-AT. Resolution of OSA after AT was defined by a post-AT apnea-hypopnea index (AHI) < 1.5/h total sleep time (TST). Residual OSA was defined as a post-AT AHI > 5/h TST, which is considered clinically significant.

RESULTS

AT significantly improved the AHI from 15.9 ± 16.4 to 4.1 ± 5.3/h TST in 182 children (P < 0.001). Of 182 children, residual OSA (post-AT AHI > 5) was seen in 46 children (25%). Among children who had hsCRP levels measured pre- and post-AT (n = 155), mean hsCRP levels pre-AT were 0.98 ± 1.91 mg/L and were significantly reduced post-AT (0.63 ± 2.24 mg/dL; P = 0.011). Stratification into post-AT AHI groups corresponding to < 1.5/h TST, 1.5/h TST < AHI < 5/h TST, and AHI > 5/h TST revealed post-AT hsCRP levels of 0.09 ± 0.12, 0.57 ± 2.28, and 1.49 ± 3.34 mg/L with statistical significance emerging comparing residual AHI > 5/h TST compared to post-AT AHI < 1.5/h TST (P = 0.006). Hierarchical multivariate modeling confirmed that pre-AT AHI and post-AT hsCRP levels were most significantly associated with residual OSA.

CONCLUSIONS

Even though AT improves OSA in most children, residual OSA is frequent. Assessment of post-AT hsCRP levels emerges as a potentially useful biomarker predicting residual OSA.