The nonlinear association between albumin levels and risk of noninvasive fungal rhinosinusitis.

PURPOSE

The preoperative diagnosis of noninvasive fungal rhinosinusitis (NIFRS) is inaccurate, and biomarkers to assist the diagnosis are urgently needed. We aimed to evaluate the relationship between albumin levels and NIFRS to assist in early diagnosis.

METHODS

Patients with NIFRS and chronic sinusitis were enrolled in this study. Appropriate statistical methods were used to determine whether there was a statistical difference between the groups. Subgroup analysis was performed to investigate the relationship between albumin levels and NIFRS, and a generalised additive model (GAM) was used to perform nonlinear relationships.

RESULTS

A total of 620 patients were included, including 240 patients with NIFRS. A close relationship was found between albumin levels and NIFRS (P < 0.0001), and the low albumin group was associated with a higher incidence of NIFRS, which was reduced by 60 and 70% in the middle and high albumin groups, respectively. The subgroup analysis also demonstrated an association between albumin levels and NIFRS, except in patients with an alcohol history (P = 0.0665). Interestingly, a nonlinear relationship is observed according to the adjusted GAM. The inflection point was set at 37.0 g/L. A negative correlation was observed among patients with albumin > 37.0 g/L. When the albumin count was <37.0 g/L, the Y value obviously increased and was saturated at 70%, with no further significant increase.

CONCLUSION

Albumin levels were significantly negatively correlated with the incidence of NIFRS, and the incidence increased markedly among patients with albumin < 37.0 g/L.