Airway management is a major responsibility for the anesthesiologist. Difficulties with tracheal intubation significantly contribute to the morbidity and mortality associated with anesthesia. Identifying situations and patients at frequent risk for airway management problems is a key to optimal care and has been the focus of numerous publications (1,2).
Several reviews have reported that endotracheal intubation is more difficult in obese than in lean patients (2–7). However, this assertion remains debated because others studies have found no evidence that tracheal intubation is more difficult in obese than in lean individuals (1,8,9). One of the reasons for these discrepancies is the lack of consensus on the definition of the term “difficult intubation,” which varies between authors. However, an objective scoring system has been proposed to assess the intubation difficulty: the intubation difficulty scale (IDS) score, which has been validated (10). This score uses several variables associated with difficult intubation. Comparisons of the conditions of tracheal intubation between obese and lean patients have not been performed with this objective score. The objective of this study was to compare the incidence of difficult tracheal intubation between lean and obese patients by using the IDS score.
After IRB approval and written, informed consent were obtained, all obese (body mass index [BMI] ≥35 kg/m2) adult (older than 18 yr) patients scheduled for laparoscopic gastroplasty in our university hospital during a period of 10 mo were included in this prospective study. Concomitantly, all the lean (BMI <30 kg/m2) adult patients who were scheduled for inguinal hernia repair or laparoscopic cholecystectomy during the same period and who were intubated by the same anesthesiologists were included in the control group. Noninclusion criteria consisted of an ASA class III or IV or a BMI between 30 and 35 kg/m2.
Preoperative airway assessment was performed by an attending anesthesiologist. Five attending anesthesiologists participated in the recruitment and induction of patients.
For each patient, five variables that may predict difficult intubation were collected: (a) the modified Mallampati classification without phonation (class I: soft palate, fauces, uvula, and pillars visible; class II: soft palate, fauces, and uvula visible; class III: soft palate and base of uvula visible; and class IV: soft palate not visible) (11,12); (b) range of head and neck motion, measured as described by Wilson et al. (7) (with differentiation of two groups: <80° and ≥80°); (c) width of mouth opening, measured as the interincisor gap in centimeters with the mouth fully opened (with differentiation of two groups: <35 and ≥35 mm); (d) presence or absence of buck teeth; and (e) presence or absence of mandibular recession. Abnormalities associated with difficult laryngoscopy (e.g., malformations, airway tumor, and loose teeth) were also recorded. In addition, before surgery, all obese patients and controls underwent a clinical evaluation and all obese patients underwent a polysomnographic study to detect significant comorbidities, including snoring, obstructive sleep apnea syndrome (OSAS), and diabetes mellitus.
Each patient was routinely monitored by an electrocardioscope, pulse oximetry, noninvasive blood pressure, and measurements of end-tidal carbon dioxide and oxygen tensions in the operating room. Hydroxyzine (100 mg) was given orally as premedication 2 h before surgery. Effervescent cimetidine (800 mg) was also given in the obese patients. Before induction, the patient was placed in a semirecumbent position (30°) with the head in the sniffing position (13). A tight face mask was applied to ensure preoxygenation, which was maintained until end-tidal oxygen reached 85% (Capnomac Ultima; Datex Engström, Helsinki, Finland). The duration of preoxygenation was noted. Anesthesia was then induced with propofol (2–2.5 mg/kg) and succinylcholine (1 mg/kg), with the dosages previously recommended (14,15). Cricoid pressure was applied as described by Sellick (16). A Macintosh No. 3 laryngoscope blade was used for the first laryngoscopy in every case. The oxygen saturation (SpO2) value obtained at the end of preoxygenation and the minimal value of SpO2 measured during the intubation procedure were noted.
Visualization of the glottis during laryngoscopy was assessed with the modified Cormack classification (class I, complete visualization of the vocal cords; class II, visualization of the inferior portion of the glottis; class III, visualization of the epiglottis only; and class IV, inability to visualize the epiglottis) (17). Intubation difficulty was assessed with the IDS developed by Adnet et al. (10) on the basis of seven variables associated with difficult intubation. They are as follows: N1, number of additional intubation attempts; N2, number of additional operators; N3, number of alternative intubation techniques used; N4, glottic exposure as defined by Cormack and Lehane (17) (grade 1, N4 = 0; grade 2, N4 = 1; grade 3, N4 = 2; and grade 4, N4 = 3); N5, lifting force applied during laryngoscopy (N5 = 0 if inconsiderable and N5 = 1 if considerable, as assessed subjectively); N6, need to apply external laryngeal pressure to improve glottic pressure (N6 = 0 if no external pressure or only the Sellick maneuver was applied and N6 = 1 if external laryngeal pressure was used); and N7, position of the vocal cords at intubation (N7 = 0 if abducted or not visible and N7 = 1 if adducted). The IDS score is the sum of N1 through N7. A score of 0 indicated intubation under ideal conditions, performed on the first attempt by the first operator, who used a single technique and applied minimal force to insert the tube through a fully visualized glottis. An IDS score from 1 to 5 indicated slight difficulty, and an IDS score >5 indicated moderate to major difficulty (10). In this study, we defined two groups of patients according to the IDS values: those with an IDS score <5 (i.e., easy and slight difficulty) and those with an IDS score ≥5 (i.e., difficult intubation).
Assuming a percentage of difficult intubation (i.e., IDS ≥5, which was the primary outcome) of approximately 2%(13) in the lean patients, we calculated the appropriate sample size with use of α = 0.05 and β = 0.20. Because a previous study suggested that the incidence of difficult laryngoscopy was 13% in obese patients and that the incidence of morbidly obese patients requiring awake intubation was 8%(18,19), we postulated that the incidence of difficult tracheal intubation in obese patients would be 11%. With this assumption that obesity might increase the incidence of difficult tracheal intubation from 2% to 11%, at least 115 patients per group would be necessary. We included 140 patients per group.
We first compared the IDS values, patient characteristics, and preoxygenation data between obese and lean patients by using a univariate analysis. A χ2 test with Yates correction or a Fisher’s exact test was used for comparisons of qualitative variables. Nonparametric tests (Mann-Whitney U-tests or Kruskal-Wallis tests) were used for comparisons of quantitative variables.
In a second step, a univariate analysis was performed to determine the risk factors for difficult tracheal intubation in the obese patients alone. We compared the obese patients with an IDS score <5 and ≥5. All the significant variables in this univariate analysis were entered in a binary stepwise multivariate logistic regression (backward-Wald) model to determine independent risk factors for difficult tracheal intubation. Continuous variables were transformed into binary variables by using the median value of the population as a cutoff. Odds ratios and 95% confidence intervals were calculated.
Values are given as mean ± SD (range), number of patients, or percentages. P < 0.05 was considered statistically significant.
One-hundred-thirty-eight obese (2 patients with incomplete data) and 140 lean patients met the conditions required for evaluation in the operating room. Nine obese and six lean patients were eliminated because tracheal intubation was performed by a resident. Finally, 129 obese and 134 lean patients were included in this prospective study.
No intubation was impossible in this series. Laryngoscopies were possible for all patients. The incidence of difficult intubation was more frequent in the obese than in the lean patients; 83 (61.9%), 48 (35.8%), and 3 (2.3%) lean patients had an IDS score of 0, >1, <5, and ≥5, respectively, whereas the numbers for obese patients were 56 (43.3%), 53 (41.1%), and 20 (15.5%), respectively (P < 0.001). During the intubation procedure, the number of attempts was 1 (range, 1–4) and 1 (range, 1–8), the number of operators involved in the procedure was 1 (range, 1–3) and 1 (range, 1–4), and the number of techniques used was 1 and 1 (range, 1–6) in the lean and obese patients, respectively. Other patient characteristics are displayed in Table 1. The incidence of comorbidities was more frequent in the obese than in the lean patients. The incidence of difficult laryngoscopy (Cormack class III or IV) was similar between lean (10.4%) and obese (10.1%) patients (P = not significant). The duration of preoxygenation and the value of SpO2 at the end of preoxygenation were similar between lean and obese patients. The minimal value of SpO2 noted during the procedure was higher in lean than in obese patients (Table 1). Among the 20 obese patients for whom the IDS value was ≥5, the mean minimal value of Spo2 during the tracheal intubation procedure was 89% ± 10% (range, 50%–99%), whereas it was 96% ± 7% (range, 64%–100%) in the obese patients for whom the IDS value was <5 (P = 0.0006). The time to intubation was not recorded during the study.
A univariate analysis was then performed with the obese patients only to determine the risk factors for difficult intubation in this population. We compared obese patients with an IDS score <5 and those with an IDS score ≥5 (Table 2). A multivariate analysis was performed with the significant variables of the univariate analysis (Table 3). The multivariate analysis demonstrated that a Mallampati score of III or IV was an independent risk factor for difficult intubation in obese patients, whereas obesity (i.e., BMI) was not. The sensitivity of the Mallampati score was 100% and 85%, its specificity was 74% and 62%, its positive predictive value was 8% and 29%, and its negative predictive value was 100% and 96% in lean and obese patients, respectively.
These results indicate that difficult tracheal intubation is more frequent in obese than in lean patients. In this study, the rate of difficult intubation was 15.5% in the obese patients and 2.2% in the lean patients. The latter figure is in keeping with the 1% to 4% range found in earlier studies of nonobstetrical unselected patients (2,13). Our data agree with several review articles supporting an association between obesity and difficult intubation (2–5,7,19,20). However, this association has been challenged because the studies that demonstrated that obesity was a risk factor for difficult intubation presented methodological limitations that call into question the validity of their findings, whereas others studies demonstrated that obesity was not associated with an increased incidence of difficult intubation.
First, the studies that previously demonstrated that obesity was a risk factor for difficult intubation presented methodological limitations. The association between obesity and difficult intubation was previously found in noncomparative studies (4,19) or in studies of small numbers of patients (7,20). For instance, in a study showing that intubation was more difficult in obese than in nonobese women during delivery, the statistical analysis included only 17 and 8 patients in these 2 groups, respectively (20). Similarly, Wilson et al. (7), who identified obesity as a risk factor for difficult intubation, were able to include only two obese patients and one lean patient with intubation difficulties. It is more important to note that all these previous studies failed to distinguish between difficult intubation and difficult laryngoscopy. The two do not necessarily go together, however. For instance, in our study, intubation was more difficult in the obese patients, whereas the incidence of difficult laryngoscopy (i.e., Cormack class III or IV) was similar in obese and lean patients. This is not surprising, because factors complicating laryngoscopy do not reflect the full spectrum of complex events that can make intubation difficult or easy. The need for a clinically relevant definition of difficult intubation prompted us to use the IDS score, which improved the reliability of identifying difficult tracheal intubation (10,21). The use of the IDS score allowed us to demonstrate that tracheal intubation, not laryngoscopy, was more difficult in obese than in lean patients. Second, the negative previous studies, which suggested that obesity and weight were not risk factors for difficult intubation, also failed to distinguish between difficult intubation and difficult laryngoscopy (1,6,8,9). In addition, some of these studies were performed with a small number of patients (9), without control (i.e., lean) patients (8), or even without obese patients (1).
A multivariate analysis restricted to the obese group was conducted to look for factors predicting difficult intubation in these patients. In keeping with the results reported by Brodsky et al. (8), we found that a Mallampati score of III or IV was a risk factor for difficult intubation in obese patients. However, as previously described, the sensitivity, specificity, and negative predictive value of the Mallampati score were poor (2), and this calls into question the validity of this predictive factor in clinical practice. One can suggest that the clinical predictive value of the Mallampati score is overridden by the degree of jaw mobility, which is often limited in obese patients by simple mass effect. In agreement with Brodsky et al., we also found that the BMI was not an independent risk factor for difficult tracheal intubation in obese patients (8). In other words, among obese patients, this result suggests that the most severely overweight were not more difficult to intubate than the others. In addition, we observed that in obese patients, difficult intubation was not significantly associated with any of the other risk factors established in the general (lean) population, including those demonstrated by Wilson et al. (7) (snoring, abnormal spinal mobility, receding mandible, buck teeth, and <35 mm of mouth opening). Furthermore, OSAS, a well known risk factor for difficult laryngoscopy in lean individuals (22,23), was not associated with difficult intubation in our obese patients. The alteration of the anatomy of upper airways in the obese patients may explain these discrepancies between lean and obese. Moreover, as described previously, all the previously described risk factors were in fact risk factors for difficult laryngoscopy and not for difficult intubation.
We also observed that hypoxemia occurred more often in obese than in lean patients during anesthesia induction, despite a similar preoxygenation. These data are in agreement with the report that apnea-induced desaturation develops more rapidly in obese than in lean patients, despite preoxygenation (24). This fact is classically related to a reduction in functional residual capacity, which is usual in obese patients (3,25). This reduction of functional residual capacity is also accompanied by a decrease in compliance, an increase in airway resistance, and an increase in pulmonary vascular resistances (3). It is worth mentioning that, in this study, a desaturation occurred more frequently and was more important in the obese patients with a difficult intubation than in those without difficult intubation. This result suggests that difficult intubation is another common and important factor that contributes to the increased risk of hypoxemia in obese patients during the induction of anesthesia by increasing the time needed to insert the tracheal tube. Routine awake intubation of patients with morbid obesity has, therefore, been recommended as a means of minimizing this risk of desaturation (5,9,25). However, this aggressive approach, which has not been validated, is cumbersome and generates patient discomfort. In addition, it is unnecessary in most cases (26), as demonstrated in this study, in which tracheal intubation by direct laryngoscopy was successful in all obese patients. Thus, routine awake intubation is not mandatory in obese patients.
Our study has several limitations. The IDS score could have been intentionally increased because the anesthesiologists knew the primary purpose of this study, but it was impossible to maintain blindness of the study group. However, the fact that the same anesthetic procedure was used for all patients, the fact that the IDS was assessed by a small number of anesthesiologists, and the nature of the IDS score may have minimized the investigator bias. Another limitation of our results was the small sample used to identify the risk factors for difficult tracheal intubation in obese patients. However, studying the risk factors for difficult intubation in the obese was not the primary end-point, and the sample size was appropriate for the primary outcome. It is nevertheless a fact that more obese patients with difficult intubation are needed to exhaustively identify the risk factors for difficult intubation in this population. Finally, the Sellick maneuver has been reported to cause upper airway obstruction and more difficult intubation over difficult laryngoscopy (27). However, because the Sellick maneuver was applied in both lean and obese subjects, we suggest that its effect did not alter our conclusions.
We conclude that difficult tracheal intubation is more common among obese than nonobese patients. Among the classic risk factors for difficult intubation, only a Mallampati score of III or IV is a risk factor in obese patients. However, its value in daily practice is poor. The increased risk of desaturation during difficult intubation should be borne in mind when anesthesia is induced in obese patients. Skilled anesthetic assistance and a wide range of equipment to facilitate intubation should be available. The risk of hypoxemia and the paucity of elements predicting difficult intubation warrant studies aimed at identifying new predictors of difficult intubation in obese patients.
We thank Professor Jean Mantz, MD, PhD, for his helpful comments.
1. Karkouti K, Rose DK, Wigglesworth D, Cohen MM. Predicting difficult intubation: a multivariable analysis. Can J Anaesth 2000; 47: 730–9.
2. Benumof JL. Management of the difficult adult airway: with special emphasis on awake tracheal intubation. Anesthesiology 1991; 75: 1087–110.
3. Adams JP, Murphy PG. Obesity in anaesthesia and intensive care. Br J Anaesth 2000; 85: 91–108.
4. Fox GS, Whalley DG, Bevan DR. Anaesthesia for the morbidly obese: experience with 110 patients. Br J Anaesth 1981; 53: 811–6.
5. Shenkman Z, Shir Y, Brodsky JB. Perioperative management of the obese patient. Br J Anaesth 1993; 70: 349–59.
6. Voyagis GS, Kyriakis KP, Dimitriou V, Vrettou I. Value of oropharyngeal Mallampati classification in predicting difficult laryngoscopy among obese patients. Eur J Anaesthesiol 1998; 15: 330–4.
7. Wilson ME, Spiegelhalter D, Robertson JA, Lesser P. Predicting difficult intubation. Br J Anaesth 1988; 61: 211–6.
8. Brodsky JB, Lemmens HJ, Brock-Utne JG, et al. Morbid obesity and tracheal intubation. Anesth Analg 2002; 94: 732–6.
9. Bond A. Obesity and difficult intubation. Anaesth Intensive Care 1993; 21: 828–30.
10. Adnet F, Borron SW, Racine SX, et al. The intubation difficulty scale (IDS): proposal and evaluation of a new score characterizing the complexity of endotracheal intubation. Anesthesiology 1997; 87: 1290–7.
11. Mallampati SR, Gatt SP, Gugino LD, et al. A clinical sign to predict difficult tracheal intubation: a prospective study. Can Anaesth Soc J 1985; 32: 429–34.
12. Samsoon GL, Young JR. Difficult tracheal intubation: a retrospective study. Anaesthesia 1987; 42: 487–90.
13. Adnet F, Baillard C, Borron SW, et al. Randomized study comparing the “sniffing position” with simple head extension for laryngoscopic view in elective surgery patients. Anesthesiology 2001; 95: 836–41.
14. Rose JB, Theroux MC, Katz MS. The potency of succinylcholine in obese adolescents. Anesth Analg 2000; 90: 576–8.
15. Servin F, Pommereau R, Leresche M, et al. Use of propofol to induce and maintain general anaesthesia in morbidly obese patients. Eur J Anaesthesiol 1991; 8: 323–4.
16. Sellick BA. Cricoid pressure to control regurgitation of stomach contents during induction of anaesthesia. Lancet 1961; 2: 404–6.
17. Cormack RS, Lehane J. Difficult tracheal intubation in obstetrics. Anaesthesia 1984; 39: 1105–11.
18. Dominguez-Cherit G, Gonzalez R, Borunda D, et al. Anesthesia for morbidly obese patients. World J Surg 1998; 22: 969–73.
19. Buckley FP, Robinson NB, Simonowitz DA, Dellinger EP. Anaesthesia in the morbidly obese: a comparison of anaesthetic and analgesic regimens for upper abdominal surgery. Anaesthesia 1983; 38: 840–51.
20. Hood DD, Dewan DM. Anesthetic and obstetric outcome in morbidly obese parturients. Anesthesiology 1993; 79: 1210–8.
21. Benumof JL. Intubation difficulty scale: anticipated best use. Anesthesiology 1997; 87: 1273–4.
22. Hiremath AS, Hillman DR, James AL, et al. Relationship between difficult tracheal intubation and obstructive sleep apnoea. Br J Anaesth 1998; 80: 606–11.
23. Benumof JL. Obstructive sleep apnea in the adult obese patient: implications for airway management. J Clin Anesth 2001; 13: 144–56.
24. Jense HG, Dubin SA, Silverstein PI, O’Leary-Escolas U. Effect of obesity on safe duration of apnea in anesthetized humans. Anesth Analg 1991; 72: 89–93.
25. Berthoud MC, Peacock JE, Reilly CS. Effectiveness of preoxygenation in morbidly obese patients. Br J Anaesth 1991; 67: 464–6.
26. Powell J, Myles P, Sultana A. Obesity and difficult intubation. Anaesth Intensive Care 1994; 22: 315.
27. Allman KG. The effect of cricoid pressure application on airway patency. J Clin Anesth 1995; 7: 197–9.