Sleep endoscopy versus modified Mallampati score in sleep apnea and snoring

The aim of this study is to quantify the correlation between a large tongue classified by the modified Mallampati score (MMS) and obstruction at the tongue base observed during sleep endoscopy in patients with socially unacceptable snoring (SUS) and obstructive sleep apnea syndrome (OSAS).

Prospective.

One hundred twenty-seven patients with obstructive sleep apnea and SUS underwent sleep endoscopy. Patients were classified into four specific groups according to their MMS. After intravenous administration of midazolam, obstructive events in the upper airway were classified. The level of obstruction was expressed as palatal (2), palatal and to a lesser degree base of the tongue/laryngeal level (2), tongue base/laryngeal level and to a lesser degree palatal (3), and tongue base/laryngeal level (3). The degree of correlation between the level of obstruction and the MMS was analyzed by means of Spearman's rank-order correlation.

No linear association between the level of obstruction and MMS could be found. A large tongue (MMS 3 or 4) does not predict obstruction at the base of the tongue. Patients classified as MMS 1 or 2 did not frequently show obstruction at the base of the tongue (9 of 38, 23,6%). The majority of the population showed retropalatal obstruction. A large tongue (MMS 3 or 4) was often accompanied by retropalatal obstruction, 76% of the cases (64 of 89).

No positive correlation exists between a large tongue and obstruction at tongue base level. In both groups, retropalatal obstruction occurred more often than retrolingual obstruction. Sleep endoscopy provides qualitative insight into important changes during sleep in patients with SUS and OSAS. In addition, anatomic aberrations are clearly visualized in a dynamic setting. Classification of the size of the tongue, as assessed by the MMS, may be of great importance as well.