Head and Neck Physical Examination: Comparison Between Nonapneic and Obstructive Sleep Apnea Patients

The Laryngoscope Lippincott Williams & Wilkins, Inc. 2005 The American Laryngological, Rhinological and Otological Society, Inc. Head and Neck Physical Examination: Comparison Between Nonapneic and Obstructive Sleep Apnea Patients Adriane I. Zonato, MD, PhD; Fernanda Louise Martinho, MD; Lia Rita Bittencourt, MD, PhD; Osíris de Oliveira Camponês do Brasil, MD, PhD; Luiz Carlos Gregório, MD, PhD; Sergio Tufik, MD, PhD Presented at the 7th World Congress in Sleep Apnea, Helsink, Finland, June 30 July 3, 2003. From the Otorhinolaryngology Discipline (F.L.M., O.D.O.C.D.B., L.C.G.), Universidade Federal de São Paulo, São Paulo, Brazil; and the Sleep Institute, Psychobiology Department (A.I.Z., L.R.B., S.T.), Universidade Federal de São Paulo, São Paulo, Brazil. Editor s Note: This Manuscript was accepted for publication February 18, 2005. Send Correspondence to Dr. Adriane I. Zonato, Rodovia Jornalista Manoel de Menezes, 2031 Casa 29, Cep: 88061 701 Florianópolis-SC, Brazil. E-mail: aizonato@terra.com.br DOI: 10.1097/01.MLG.0000163494.19965.DC Study Objectives: The purpose of this study was to apply a systematic physical examination, used to evaluate obstructive sleep apnea (OSA) patients, in nonapneic patients. Design: Study was prospective. Setting: Patients were seen in the sleep laboratory and department of otorhinolaryngology. Patients or Participants: Nonapneic patients (n 100) were involved in the study. Interventions: Physical examination to evaluate facial skeleton, pharyngeal soft tissue, rhinoscopy, and body mass index. Data were compared with a previously published study (2003) on a group of OSA patients (n 223). Measurements and Results: Skeletal examination detected retrognathism in 6%, class II occlusion in 12%, and high-arched hard palate in 11%. The modified Mallampati classification showed 54% in class I to II and 46% in class III to IV. Only 1% of nonapneic patients had tonsils of degree III to IV. Oropharynx evaluation showed web palate in 38%, posterior palate in 19%, thick palate in 10%, thick uvula in 10%, long uvula 15%, voluminous lateral wall in 11%, and tongue edge crenations in 28%. Anterior rhinoscopy detected significant septal deviation in 1% and turbinate hypertrophy in 31% of patients. Conclusions: The head and neck physical examination, considering both skeletal and soft tissue alterations, illustrated significant differences between nonapneic and OSA patients. Body mass index, modified Mallampati classification, tonsils hypertrophy, and high-arched hard palate previously related to the presence of sleep apnea in the literature showed different outcomes in nonapneic patients. Nonapneic patients had less alterations in nasal anatomy (severe septal deviation and enlarged turbinate). Skeletal parameters, such as retropositioned mandible and angle class II occlusion, were less frequent in nonapneic patients. Key Words: Sleep apnea, physiopathology, physical examination, anatomical abnormalities, nasal obstruction. Laryngoscope, 115:1030 1034, 2005 INTRODUCTION Although obstructive sleep apnea (OSA) has been described, a complete understanding of the physiopathology and upper airway (UAW) anatomic participation has yet to be reached. However, the scientific community has reached a consensus on how the occurrence of both anatomic abnormality and neuromuscular disorders can contribute to or facilitate the narrowing and closure of the pharynx during sleep. 1,2 Many authors have shown that relevant physical alterations (craniofacial, dental-bite, pharyngeal soft tissue, UAW size and shape, obesity) can easily be recognized in patients with OSA. 3 8 Despite such anatomic evidence, the relationship between anatomic abnormalities and the presence and severity of sleep apnea is unclear. Some physical aspects have been related to the presence or severity of OSA. These aspects are body mass index (BMI), modified Mallampati classification (MMC) (assesses palatal length and tongue size), tonsils size, low hyoid bone position, and high-arched hard palate. 6 9 None of the usual anatomic alterations seen in the soft palate (posterior placed, thick, or webbed) or in the uvula (thick or long) had any correlation to the presence or severity of sleep apnea. Zonato et al. 8 stated that at least one of the above cited soft-palate or uvula alterations were seen in 34% to 45% of the reviewed patients. Despite this, none of the palatal findings alone highlighted any correlation with the presence of sleep apnea. The authors observed that only pharyngeal abnormalities in combination (the presence of at least 3 of these 5 pharyngeal characteristics: tonsil size grade III or IV, abnormal palate, abnormal uvula, voluminous lateral wall, or web palate) were predictive of both presence and severity of OSA. 1030

Fig. 1. Modified Mallampati classification (the relative position of the palate and base of tongue inside the mouth). Class I: all the oropharynx including tonsils, pillars, soft palate, and the tip of uvula can be easily visible. Class II: tonsils upper polo and uvula are visible. Class III: part of the uvula and soft palate are visible. Class IV: just the hard palate and part of soft palate are visible. There is no consensus concerning the observed physical alterations that are important in patients with OSA or those that contribute to the physiopathology of sleep apnea. Only anthropometric measurements (BMI, neck and abdominal circumference) and MMC are seen to be the common denominators among the published studies so far. Furthermore, there are some patients with sleep apnea who do not present the most common physical abnormalities. As far as another group of patients without OSA is concerned, a question arises: would they present these most commonly described anatomic alterations? The objective of the present study was to apply a systematic head and neck physical examination in nonapneic patients to compare them with OSA patients. 8 MATERIALS AND METHODS Study Design We carried out a prospective study involving 100 patients from the Sleep Laboratory of Universidade Federal de São Paulo, during the period of April to September 2003. The patients were seen in the nonrespiratory sleep disturbance out-patient unit or on the evening before the polysomnography (PSG) study. Inclusion criteria comprised adult patients evaluated with an overnight standard PSG at the Sleep Institute, with an apneahypopnea index (AHI) of 5 or less events/hour, without habitual snoring. Exclusion criteria were patients with AHI greater than 5 events/hour, patients with AHI 5 or less events/hour with habitual snoring. Snoring was classified as habitual if it was present every night or almost every night. All subjects were questioned about past history of hypertension, cardiovascular diseases, diabetes mellitus, hypothyroidism, nasal obstruction, and rhinitis. PSG was performed for one full night using an SAC Oxford system containing 13 channels including electroencephalogram, electrocardiogram, oculogram, submental and anterior-tibialis electromyogram, nasal canula and oral airflow, thoracic and abdominal movements, body position, and oxygen saturation measured by pulse oximetry. The subjects underwent a specific otolaryngologic examination by the same ear, nose and throat investigator. All the physical characteristics of the nonapneic group (n 100) were compared with an OSA group of patients (n 223) whose results had already been published by our group. 8 Physical evaluation was assessed by a systematic head and neck examination, comprised of the following evaluation. 8 Facial Morphology Presence of mandible retrognathism is investigated. The patient is seated with the head in the Frankfort horizontal position, and a virtual vertical line is dropped from the vermilion border of the lower lip to the chin. If the anterior prominence of the chin (soft tissue pogonion) is greater than 2 mm behind this line, mandibular retro-displacement may be present. 10 Oral Cavity Oral cavity examination starts by looking for dental occlusion Angle Class II (possibility of mandibular retrognathism). By asking the patient to open his mouth with the tongue relaxed inside the mouth, the MMC is graded according to palatal length and tongue size as class I (all the oropharynx including tonsils, pillars, soft palate, and the tip of uvula can be easily visible), class II (tonsils upper polo and the uvula are visible), class III (part of the uvula and soft palate are visible), and class IV (just the hard palate and part of soft palate are visible) 7 (Fig. 1). The skeletal structure of the hard palate is inspected with attention to the presence of high and narrow arched palate (Fig. 2). The tongue inspection looks for the presence of edge crenations (consequence of teeth pressure) as a subjective sign of large tongue. Oropharynx In the oropharynx examination, soft palate is classified as posterior placed, thick (wide aspect of the tissue), and webbed (web palate caused by redundant posterior pillars); the lateral wall as voluminous with wrinkling pillars; and the uvula as thick (wide aspect of the tissue) and long (greater than 1 cm). Tonsils are classified as either degree I, II, III, or IV according to its hypertrophy (I: tonsils inside the tonsillar fossa lateral to posterior pillars; II: tonsils occupy 25% of oropharynx; III: tonsils occupy 50%; IV: tonsils occupy 75% or more, almost meeting in the midline) and as degree 0 (previous tonsillectomy). Nasal Examination Nasal examination by anterior rhinoscopy is used to detect septal deviation and turbinate hypertrophy. Fig. 2. High-arched hard palate. 1031

TABLE I. Nonapneic Patients and OSA Patients: Clinical Aspects. Nonapneic Patients (n 100) OSA Patients (n 223) P Value Male:Female (%) 46:54 64:36.002 Age 44.6 48.039 BMI 24.5 29.001 Hypertension (%) 22 31.8 NS Diabetes mellitus (%) 6 7.2 NS Cardiovascular disease (%) 9 5.8 NS Hypothyroidism (%) 4 4.9 NS Nasal obstruction (%) 42 64.1.001 Rhinitis (%) 39 40.4 NS OSA obstructive sleep apnea patients; BMI body mass index. Statistical Analysis Statistical differences between nonapneic and OSA patients were determined by the Mann-Whitney U test or the chi-square test, depending on the type of the variable. For all tests, the level of significance was set for P.05. Our data were analyzed using software Statistica for Windows release 5.1 (Statsoft, Inc., tulsa, OK). RESULTS One hundred patients were included in this analysis with a mean age of 44.6 13.8 (range 18 70) years, with 46 (46%) men and 54 (54%) women and BMI of 24 3.9 (range 16.8 35.4) kg/m 2 (Table I). On clinical evaluation, hypertension was present in 22 (22%) patients, cardiovascular disease in 9 (9%), diabetes mellitus in 6 (6%), and hypothyroidism in 4 (4%) patients. In addition, nasal obstruction was seen in 42 (42%) patients and symptoms of rhinitis in 39 (39%) patients. All 100 patients were recorded by standard polysomnogram with a mean AHI of 2.8 1.8 (range 0 5) events/hour (Table II). Facial skeletal examination detected retrognathism in six (6%) patients. Evaluation of dental occlusion showed that class II was observed in only 12 (12%) patients. Some patients (n 38) were totally or partially edentate, preventing a real evaluation of the occlusion. The MMC showed 20 patients (20%) in class I, 34 (34%) in class II, 31 (31%) in class III, and 15 (15%) in class IV. Classes III and IV of MMC, considered a sign of difficult tracheal intubation and small mouth, was observed in 46 (46%) patients. High-arched hard palate was observed in 11 (11%) patients and tongue edge crenations by teeth in 28 (28%). The oropharynx evaluation showed web palate in 38 patients (38%), posterior placed palate in 19 (19%), thick soft palate in 10 (10%), thick uvula in 10 (10%), long uvula 15 (15%), and voluminous lateral wall with wrinkling pillars in 11 (11%). The score of tonsils size was grade 0 24 (24%); grade I 53 (53%); grade II 22 (22%); and grade III 1 (1%), with no patient in grade IV. Taking patients with a low degree of tonsil hypertrophy (grades I and II) along with those who had previous tonsillectomy, we saw that they comprised almost the entire group (99%). Physical examination by anterior rhinoscopy detected some degree of septal deviation in 49 (49%) patients (degree I 22%; degree II 26%; degree III 1%) and turbinate hypertrophy in 31 (31%) patients. All the investigated physical variables were compared with the OSA group, and the findings with statistical significance are listed in Table III. DISCUSSION According to the literature, the anatomic abnormalities that have a correlation with sleep apnea are anthropometric measures (BMI, abdominal and neck circumference), MMC, high-arched hard palate, pharyngeal anatomic abnormalities in combination (presence of at least 3 abnormalities such as hypertrophied tonsils, abnormal palate, abnormal uvula, voluminous lateral wall, or web palate), low hyoid bone position (in this case, using complementary examination by cephalometric analysis), posterior wall redundancy, and endoscopy retropalatal size (analyzed by videotape endoscopy and lateral cephalometric radiograph). 4,6 9,11 13 In a previous paper, we observed that many of the anatomic parameters were frequently seen in OSA patients (retropositioned mandible was detected in 19.7%, class II occlusion in 26.3%, tongue edge crenation in 36.3%, web palate in 45.3%, posterior placed palate in 43%, thick palate in 35%, long uvula in 34.1%, thick uvula in 34.5%), but the statistical analyses failed to highlight any correlation between these parameters and AHI. According to MEDLINE data, we found only two studies that compared nonapneic individuals with OSA patients. Viner et al. 14 used a combination of history and physical examination as a screening for sleep apnea in 410 patients referred for suspected sleep apnea syndrome. All patients underwent clinical and physical examination with a subsequent PSG study, and the group was divided into patients with (AHI 10) and without sleep apnea. The UAW anatomic evaluation was limited; the only investigated parameter was the examination of the pharynx, which was considered as normal or abnormal if it appeared narrow and small; if the uvula were bulky, long, or rested on the base of tongue during phonation; or if the TABLE II. Nonapneic Patients and OSA Patients: Polysomnography Data. Sleep Efficiency AHI Sleep Stage 1 Sleep Stage 2 Sleep Stage 3/4 REM Sleep Lowest Oxygen Saturation Nonapneic patients 74.8% 2.81 9.2% 56.7% 19.6% 15.6% 89% OSA patients 79.4% 23.8 10.9% 58% 15.6% 15.9% 78% P value.03.001 NS NS.003 NS.001 OSA obstructive sleep apnea patients; AHI apnea-hypopnea index; REM rapid eye movement. 1032

TABLE III. Comparison of Physical Examination Variables between Nonapneic Patients and OSA Patients. Nonapneic Patients (%) (n 100) OSA Patients (%) (n 223) P Value High-arched hard palate 11 25.1.003 Retropositioned mandible 6 19.7.001 Class II occlusion 12 26.3.005 Total or partial edentate 38 29.7 NS Tongue edge crenation 28 36.3 NS Web palate 38 45.3 NS Posterior palate 19 43.001 Thick palate 10 35.001 Long uvula 15 34.1.001 Thick uvula 10 34.5.001 Voluminous lateral wall 11 30.9.001 MMC classes I and II 54 21.2.001 MMC classes III and IV 46 78.8.001 Previous tonsillectomy 24 20.4 NS Tonsils degree I and II 75 65.2.001 Tonsils degree III and IV 1 14.4.040 Septal deviation degree I 22 30.9 NS Septal deviation degree II 26 30.9 NS Septal deviation degree III 1 5.8.048 Turbinate hypertrophy 31 49.8.001 OSA obstructive sleep apnea patients; MMC modified Malampatti classification. tonsils were big enough to fit the pharyngeal orifice. The authors observed a statistical difference between patients with and without sleep apnea: 78 of 218 patients without sleep apnea also presented an abnormal pharynx during examination, in contrast with 97 of 190 patients with sleep apnea. With sleep apnea clinical variables and pharyngeal examination to detect and differentiate patients with and without sleep apnea, the subjective impression had a sensitivity of 52% and a specificity of 70%. Deegan and McNicholas 15 carried out a study with 250 patients suspected of having sleep disordered breathing to evaluate the predictive value of clinical features to OSA. The clinical features were investigated by questionnaire, anthropometric measures (BMI, neck and abdominal circumference), and UAW examination. Only the pharynx described with redundant tissue or small, septal deviation and glossomegaly were assessed. For these three facial parameters, there was no statistical difference between the patients with and without OSA, but there was a correlation between the anthropometric measurements and the AHI. The authors considered as a non-osa group all patients with an AHI less than 15 events/hour. According to current classification, this would also comprise the mild apnea patients. This situation is also observed in Viner et al. s study, which considered patients without sleep apnea as those where AHI was 10 or greater. Neither of these previous studies had used a complete physical examination procedure for facial and UAW findings. In the present study, the majority of selected patients came from an insomnia clinic. Because of this fact, there was no prevalence of male over female patients and patients were not as obese as OSA patients commonly are. Our clinical investigation identified nasal obstruction as a clinical complaint that was more prevalent in OSA patients. In addition, we observed that the OSA group had more turbinate hypertrophy and septal deviation (only for deviation degree III, which represents the highest degree of deviation) than the nonapneic group. All the craniofacial characteristics (retropositioned mandible, class II occlusion, and high-arched hard palate) were less frequent in the nonapneic group. Despite this, our previous paper failed to observe any correlation between these skeletal parameters and AHI. The previous and actual papers data point to the fact that craniofacial evaluation appears to be relevant by presenting different expressions in OSA patients compared with nonapneic patients. We speculate that these deformities should not be ignored. The evaluation should be more precise, using objective parameters such as cephalometric measurements. All the soft tissue in the oropharynx, with the exception of web palate and tongue edge crenation, was found to be more predominant in the OSA group, with a very significant statistical value. The tongue edge crenation parameter, as an indirect sign of excessive tongue volume, was observed in both groups without any difference. This parameter is perhaps not the most adequate one to determine, subjectively, the tongue volume. The reference to the web palate was not a sound parameter. In fact, the web palate could be graduated whether the web mucosa was small or really large, with the mucosa insertion very close to the tip of uvula. Subsequently, we set out to quantify the web palate (small, moderate, and large web 1033

palate), in a new study, according to the point of mucosa insertion close to the base, middle, or the tip of uvula, and then repeated the examination in nonapneic and OSA patients. The anatomic parameters of tonsils size and oropharynx size were very interesting. The majority of nonapneic patients have tonsils of degrees I and II (75% of all group), only one patient had tonsil degree III, and none presented with severe hypertrophy. In our previous paper, we observed that just a few OSA patients had real hypertrophied tonsils (degree III or IV), in essence, less than 15% of them. In consideration of patients with previous tonsillectomy, nonapneic and OSA patients were similar. The III and IV scores MMC were present in most OSA patients (78.8%), indicating that most OSA patients may have a small oropharynx and a difficult endotracheal intubation. For our nonapenic patients, MMC class III or IV were observed in 46% of the group. This situation concurred with the data of our previous study and with Friedman et al. s 7 study where there was a correlation between the presence and severity of sleep apnea with MMC, and that this anatomic characteristic may be relevant in OSA patients. 8 For each of all other soft-tissue parameters, posterior or thick palate, long or thick uvula and voluminous lateral wall, the groups were different. Given the patient population of the present study, where the majority was referred to us by an insomnia clinic, the apneic and nonapneic groups were not well matched for sex, age, and obesity. Despite the imposed limitations of this present study, the results illustrated a real contrast in the common anatomic findings between apneic and nonapneic groups. CONCLUSION A systematic head and neck physical examination, which considered both skeletal and soft tissue alterations, illustrated significant differences between nonapneic and OSA groups. BMI, MMC, tonsil hypertrophy, and higharched hard palate arguably have different expressions in nonapneic patients. All skeletal alterations (high-arched hard palate, retroposition mandible, and class II occlusion) are less frequent in the nonapneic group. Nonapneic patients scarcely present enlarged tonsils (degree III and IV encompassed a mere 1% of patients), have fewer alterations in palate and uvula, and less nasal alterations (severe septal deviation and enlarged inferior turbinate). Significant anatomic differences were observed in the present study between apneic and nonapneic groups, but studies involving control matching clinical investigation are still needed for additional insights. The significance of anatomic findings in this field of study warrants further research and clinical investigation. BIBLIOGRAPHY 1. Isono S, Remmers JE. Anatomy and physiology of upper airway obstruction. In: Krygger MH, Roth T, Dement WC, eds. Principles and Practice of Sleep Medicine, 2nd ed. Philadelphia: W.B. Saunders, 1994:642 656. 2. Sher AE. Obstructive sleep apnea syndrome: a complex disorder of the upper air way. Otolaryngol Clin North Am 1990;23:593 608. 3. Woodson BT. Examination of the upper airway. Oral Maxillofac Surg Clin North Am 1995;7:257 267. 4. Deegan PC, McNicholas WT. Predictive value of clinical features for the obstructive sleep apnoea syndrome. Eur Respir J 1996;9:117 124. 5. Woodson BT. Predicting which patients will benefit from surgery for obstructive sleep apnea: the ENT exam. ENT J 1999;78:796 800. 6. Woodson BT, Naganuma H. Comparison of methods of airway evaluation in obstructive sleep apnea syndrome. Otolaryngol Head Neck Surg 1999;120:460 463. 7. Friedman M. Tanyeri H, La Rosa M, et al. Clinical predictors of obstructive sleep apnea. Laryngoscope 1999;109: 1901 1907. 8. Zonato AI, Bittencourt LRA, Martinho FL, et al. Association of systematic head and neck physical examination with severity of obstructive sleep apnea-hypopnea syndrome. Laryngoscope 2003;113:973 980. 9. Petri N, Suadicani P, Wildschiodtz G, Bjorn-Jorgensen J. Predictive value of Müller maneuver, cephalometric and clinical features of the outcome of uvulopalatopharyngoplasty. Evaluation of predictive factors using discriminate analysis in 30 sleep apnea patients. Acta Otolaryngol (Stockh) 1994;114:565 571. 10. Ridley MB. Aesthetic facial proportions. In: Papel ID, Nachlis NE, eds. Facial Plastic and Reconstructive Surgery. St. Louis: Mosby-Yearbook, 1992:106. 11. Teculescu DB, Montaut-Verient B, Hannhart B, et al. Breathing pauses during sleep: can a non-invasive ENT examination help identify subjects at risk in epidemiological settings? Med Hypoth 2001;56:653 656. 12. Roumbaux Ph, Bertrand B, Boudewyns A, et al. Standard ENT clinical evaluation of sleep-disordered breathing patient; a consensus report. Acta Otorhinolaryngol Belg 2002; 56:127 137. 13. Tsai WH, Remmers JE, Brant R, et al. A decision rule for diagnostic testing in obstructive sleep apnea. Am J Crit Care Med 2003;167:1427 1432. 14. Viner S, Szalai JP, Hoffstein V. Are history and physical examination a good screening test for sleep apnea? Ann Intern Med 1991;115:356 359. 15. Deegan PC, McNicholas WT. Predictive value of clinical features for the obstructive sleep apnoea syndrome. Eur Respir J 1996;9:117 124. 1034