Objective. To observe the changes of serum adiponectin (AP) levels in patients with obstructive sleep apnea hypopnea syndrome (OSAHS) and explore the correlation between serum AP and polysomnography (PSG) parameters in patients with OSAHS. Methods. The data of subjects who underwent PSG at the hospital between January 2021 and December 2022 were collected retrospectively and divided into simple snoring group (AHI < 5 times/h, n = 45), mild OSAHS group (5 ≤ AHI < 15 times/h, n = 63), moderate OSAHS group (15 ≤ AHI ≤ 30 times/h, n = 52), and severe OSAHS group (AHI > 30 times/h, n = 60). The general data, PSG indices, and serological indices of the subjects were collected and compared between groups. Pearson correlation analysis and partial correlation analysis were employed to examine the correlation between serum AP level and PSG parameters. Ordered logistic regression was employed to analyze the risk factors influencing the severity of OSAHS. The predictive capability of the serum AP level in determining the occurrence of OSAHS was assessed using ROC. The serum AP levels of subjects with different subtypes of PSG indicators were compared. Results. In the simple snoring group, mild OSAHS group, moderate OSAHS group, and severe OSAHS group, there were statistically significant differences in microarousal count, MAI, AHI, times of blood oxygen decreased by ≥ 3%, L-SaO2, and TS90% among the 4 groups (P<0.05). The level of serum AP was positively correlated with L-SaO2 and negatively correlated with the proportion of REM, microarousal count, MAI, AHI, times of blood oxygen decreased by ≥ 3%, TS90%, and LP (P<0.05). High AHI was a risk factor affecting the severity of OSAHS (95% CI: 1.446–4.170). The AUC of serum AP level in diagnosing OSAHS was 0.906 (95% CI: 0.8601–0.9521), and when the Youden Index was 0.678, the sensitivity was 88.9%, and the specificity was 78.9% (P<0.0001). In the population with a high microarousal count, high AHI, and high times of blood oxygen decreased by ≥ 3% and high TS90%, the serum AP level was lower than that in the low-level population (P<0.05). In the population with high L-SaO2, the serum AP level was higher than that in low-level population (P<0.05). Conclusion. The level of serum AP decreased with the increase of the disease severity of patients with OSAHS and demonstrates a significant predictive capability for the occurrence of OSAHS. Monitoring the level of serum AP can effectively forecast the risk of OSAHS. Furthermore, alterations in serum AP levels are associated with both hypoxemia and a heightened frequency of arousal in patients with OSAHS.