Remember me
ORIGINAL ARTICLE
Year : 2023 | Volume
: 22
| Issue : 4 | Page : 520-525
Preoperative STOP-BANG questionnaire to predict difficult airway in undiagnosed obstructive sleep apnea patients undergoing elective gynecological surgeries under general endotracheal anesthesia: A prospective observational study
Srinivas Hebbal Thammaiah1, Rashmi Hosahalli Sreenath2, Girish Bandigowdanahalli Kumararadhya1, Nivetha Babu1, KN Archana1
1 Department of Anaesthesiology and Critical Care, JSS Medical College and Hospital, JSS Academy of Higher Education and Research, Mysore, Karnataka, India
2 Department of OBG, JSS Medical College and Hospital, JSS Academy of Higher Education and Research, Mysore, Karnataka, India
Correspondence Address:
Girish Bandigowdanahalli Kumararadhya
Department of Anaesthesiology and Critical Care, JSS Medical College and Hospital, JSS Academy of Higher Education and Research, Mysore, Karnataka
India
Source of Support: None, Conflict of Interest: None
CheckDOI: 10.4103/aam.aam_2_23
Abstract
Background: STOP-BANG questionnaire is an established tool for obstructive sleep apnea (OSA) screening. Its utility in predicting difficult airway has not been evaluated. We intend to assess difficulty in airway management and associated perioperative complications in patients with undiagnosed OSA using the STOP-BANG questionnaire. Materials and Methods: We performed a prospective observational study on 250 patients with the American Society of Anesthesiologists Physical Status Class I-II, aged 18–65 years STOP-BANG score were recorded. Occurrences of difficult mask ventilation (DMV), laryngoscopy, and intubation were assessed using mask ventilation grade, intubation difficulty score intubation difficult score (IDS), and modified Cormack–Lehane grading during induction of anesthesia. Hypoxic events, hemodynamic disturbances, laryngospasm, and bronchospasm were also recorded. Results: Overall, 250 patients completed the study (Group H: n = 102, with STOP-BANG questionnaire ≥3 and Group L: n = 148, with STOP-BANG criteria < 3). A total of 67 (26.8%) cases of DMV and 63 (25.2%) cases of difficult tracheal intubation (DIT) were encountered overall. The proportion of patients with DMV during induction was 59.8% in Group H versus 4.05% in Group L (P < 0.001). A higher incidence of difficult intubation was observed in Group H (56.9% vs. 11.5%, P < 0.001). More patients in Group H had airway complications such as bleeding and injury to the posterior pharyngeal wall or teeth (P < 0.001). Complications such as hypoxia, hypertension, and tachycardia were observed to be higher in Group H (P < 0.001). Conclusion: STOP-BANG questionnaire is an effective bedside preoperative tool that helps in identifying unanticipated difficult airway.
Abstract in French Résumé
Contexte: Le questionnaire STOP-BANG est un outil établi pour le dépistage de l'apnée obstructive du sommeil (AOS). Son utilité pour prédire les voies respiratoires n'a pas été évaluée. Nous avons l'intention d'évaluer les difficultés de gestion des voies respiratoires et les complications périopératoires associées chez les patients souffrant d'AOS non diagnostiquée à l'aide du questionnaire STOP-BANG. Matériels et méthodes: Nous avons réalisé une étude observationnelle prospective sur 250 patients avec le score STOP-BANG de classe I-II de l'American Society of Anesthesiologists, âgés de 18 à 65 ans ont été enregistrés. Les occurrences de ventilation au masque difficile (DMV), de laryngoscopie et d'intubation ont été évaluées en utilisant le grade de ventilation au masque, la difficulté d'intubation score (IDS) et le classement de Cormack-Lehane modifié lors de l'induction de l'anesthésie. Événements hypoxiques, troubles hémodynamiques, laryngospasme, et le bronchospasme ont également été enregistrés. Résultats: Au total, 250 patients ont terminé l'étude (Groupe H : n = 102, avec questionnaire STOP-BANG ≥3 et Groupe L : n = 148, avec critère STOP-BANG < 3). Un total de 67 (26,8%) cas de DMV et 63 (25,2%) cas de DIT ont été rencontrés global. La proportion de patients avec DMV lors de l'induction était de 59,8 % dans le groupe H contre 4,05 % dans le groupe L (P < 0,001). Une incidence plus élevée d'intubation difficile a été observée dans le groupe H (56,9 % contre 11,5 %, P < 0,001). Plus de patients du groupe H ont eu des complications des voies respiratoires telles que saignement et lésion de la paroi postérieure du pharynx ou des dents (P < 0,001). Des complications telles que l'hypoxie, l'hypertension et la tachycardie ont été observé comme étant plus élevé dans le groupe H (P < 0,001). Conclusion: Le questionnaire STOP-BANG est un outil préopératoire efficace au chevet du patient qui aide dans l'identification des voies respiratoires difficiles imprévues.
Mots-clés: Intubation difficile, ventilation difficile du masque, score de difficulté d'intubation, apnée obstructive du sommeil, questionnaire STOP-BANG
Keywords: Difficult intubation, difficult mask ventilation, intubation difficulty score, obstructive sleep apnea, STOP-BANG questionnaire
Introduction 
Airway securement is a primary skill needed for an anesthesiologist. Difficult airway scenarios in everyday practice pose a threat to the patient's safety, especially when unanticipated. Since intubation was introduced, various airway assessment techniques and parameters have been studied in detail to predict difficult airways preoperatively.
Obstructive sleep apnea (OSA) and OSA syndrome (OSAS) have a strong association with difficult airways. OSA patients under anesthesia tend to develop more complications due to anatomical and physiological changes in the upper airway. These patients can desaturate at a much faster rate than their non-OSA counterparts. Diagnosing OSA and airway difficulty in the preoperative clinic is essential for the anesthesiologist to be better prepared to deal with a difficult airway.
The STOP-BANG Questionnaire was first developed in 2008 by Chung et al.[1] was originally validated to screen OSA in surgical patients. However, whether this questionnaire can be utilized for detecting difficult airways is unknown.
Hence, we hypothesized that the STOP-BANG questionnaire is a valuable tool in predicting the occurrence of difficult airway. If confirmed, the results of this study would provide evidence for an easier, reliable, cost-effective method that can be performed bedside to predict difficult airway in patients. The primary aim of this study was to identify difficult airway-difficult mask ventilation (DMV), difficult laryngoscopy (DL), and difficult tracheal intubation (DIT) using the STOP-BANG questionnaire. The secondary aims of this study were to evaluate the incidence of perioperative complications during induction – hypoxic events, tachycardia, bradycardia, airway trauma, dysrhythmias, laryngospasm, bronchospasm, and identification of OSA in undiagnosed patients.
Materials and Methods Study design
The study was a prospective, observational study conducted in a tertiary care teaching hospital for 1 year. Patients of age 18–65 years with the American Society of Anaesthesiologists Physical status classification I and II and scheduled for elective surgeries requiring general anesthesia and tracheal intubation were enrolled. Patients with diagnosed OSA, facial abnormalities, history of previous difficult airway, diagnosed lung disease, patients on continuous positive airway pressure therapy, saturation <90% in room air, neuromuscular disease, and pregnant patients were excluded from the study.
Study approval and trial registration
The Institutional Ethics Committee approved the study (JSS Academy of Higher Education And Research, Mysuru, JSS/MC/PG/4623/2018-19; date of approval November 2, 2018), and written informed consent was obtained from each participant. The study was registered prospectively in the Clinical Trials Registry of India (CTRI) (trial registration number: CTRI/2021/02/031339; Principal Investigator: Dr. Srinivas HT; study start date March 1, 2021), and the procedures were conducted in accordance with the Helsinki Declaration 2013.
Randomization and allocation concealment
All the patients were examined a day before surgery. Using STOP-BANG questionnaire, the patients were allocated into two groups (Group H and L).
High-risk group (Group H) – patients scoring ≥3 in STOP-BANG questionnaireLow-risk group (Group L) – patients scoring <3 in STOP-BANG questionnaire.The preanesthetic evaluation and anesthetic management in the operating room were done by two different anesthesiologists. Hence, the anesthesiologist in charge of each patient was blinded to the study group allocation and recorded perioperative data. To avoid airway difficulty due to inadequate expertise, anesthesiologists who had conducted laryngoscopy and intubation for a minimum of 100 patients were assigned to the subjects.
Anesthesia protocol
Anthropometric data including height, weight, and body mass index (BMI) were recorded. Airway assessment was done during preanaesthetic evaluation which included interincisor gap, Mallampati grade, neck circumference, sternomental and thyromental distance, and neck movements.
Monitoring, anesthesia, fluid, and temperature management were standardized for both groups. On the day of surgery, intravenous access was secured using an 18G cannula. Oxygen saturation, electrocardiogram, noninvasive blood pressure, and capnogram were monitored. Baseline parameters were recorded at the start of the procedure.
All patients were premedicated with intravenous injection midazolam 0.1 mg/kg, injection fentanyl 2 μg/kg, and lidocaine 1 mg/kg 90 s before induction and were preoxygenated for 3 min. Anesthesia was induced with injection propofol 2 mg/kg. During mask ventilation, the ease of mask ventilation, use of airway adjuncts, single or two-person technique, and any hypoxic events during this phase were noted. Injection succinylcholine 1.5 mg/kg was administered to facilitate endotracheal intubation. Laryngoscopy was performed using Macintosh blade 3 or 4 followed by intubation with an appropriate size endotracheal tube. The Cormack–Lehane (CL) grading and the use of additional equipment or maneuvers such as stylet, bougie, and external laryngeal manipulation were recorded. A maximum of three intubation attempts were allowed. A senior anesthesiologist took over and airway was managed according to difficult airway society guidelines in case of failure to intubate.[2] Mc Coy laryngoscope, C-MAC® video laryngoscope, supraglottic airway devices, or flexible fiber-optic bronchoscope were used for securing airway when needed and noted down. Any incidence of desaturation with SaO2 <95% was also noted down.
Neuromuscular blockade was maintained by injection vecuronium 0.05 mg/kg initial dosage followed by 0.02 mg/kg intermittent doses as and when required. Once the surgery was completed, neuromuscular blockade was reversed with injection neostigmine 0.05 mg/kg and injection glycopyrrolate 0.01 mg/kg after ensuring adequate recovery from neuromuscular blockade. Oral cavity and throat were thoroughly suctioned before extubation.
The occurrence of DMV and difficult intubation (DIT) was assessed using DMV score[3],[4] and intubation difficulty scale (IDS),[5] respectively. Hypoxic events if any were duly addressed and documented. Airway trauma such as oropharyngeal mucosal bleeding and injury to teeth or lip were noted. Intraoperative complications such as hypotension, hypertension, bronchospasm, laryngospasm, and dysrhythmias were also recorded.
Primary aim and outcome parameters
The primary aim of this study was to identify difficult airway-DMV, DL, and difficult tracheal intubation (DIT) using the STOP-BANG questionnaire. The primary outcome was the DMV score, Cormack–Lehane grade, and intubation difficulty score (IDS) given by Adnet et al.[5]
Secondary aims
The secondary aims of this study were to compare the incidence of perioperative complications during induction – hypoxic events, tachycardia, bradycardia, airway trauma, arrhythmias, laryngospasm, and bronchospasm among the groups and identification of OSA in undiagnosed patients.
Statistical analysis
A sample size of 206 was derived using Cochran's formula : n=Z2pq/d2:
n-Sample sizeZ-Standard scorep-Prevalenceq-(1 − p)d-precision limit.n = 1.96 × 1.96 × 0.16 × 0.84/0.05 × 0.05 = 206−250
The sample size was powered at 80%, allowing for an alpha error of 0.05, and a dropout rate of 10% was considered. We estimated our sample size to be 250 patients considering dropouts.
The Shapiro–Wilk test was used to assess the normality of the data. The parametric data were analyzed using Analysis of variance and Fisher's exact test for count data, whereas the nonparametric data were analyzed using the Kruskal–Wallis rank sum test. Categorical data were analyzed using the Chi-square test. Regression techniques were used to determine the association between the variables. Statistical significance was set at P < 0.05. Statistical data were collected in Microsoft Excel format and analyzed using R Studio version 1.3.2 (Microsoft,Washington DC, USA).
Results A total of 1040 patients were assessed for eligibility over a 1-year period, 250 of which completed the study. A total of 102 patients were included in Group H and 148 in Group L.
Baseline characteristics, including age, weight, and BMI were significantly higher in Group H (P < 0.001) [Table 1]. It was noted that the mean age was significantly higher in Group H (54 ± 11.52 vs. 41.01 ± 13.52, P < 0.001). Weight distribution was also significantly higher in Group H than Group L (80.8 ± 11.13 vs. 66.67 ± 9.98, P < 0.001), whereas height was comparable between both the groups (P = 0.0392).
Airway parameters including mouth opening given by interincisor gap, thyromental distance, sternomental distance, Mallampati grade [Figure 1], and neck circumference were observed to be higher in Group H than Group L; P < 0.001 [Table 2].
Primary outcomes
Difficult mask ventilation
The proportion of patients with DMV during induction was 59.8% in Group H versus 4.05% in Group L (P < 0.001).
In patients belonging to Group H, mask ventilation Grade 1 was seen in 41 (40.19%) cases, Grade 2 mask ventilation was seen in 24 (23.52%) patients, and Grade 3 was seen in 37 (36.27%) patients. In Group L, Grade 1 was seen in 142 (95.94%) cases, Grade 2 was seen in only 5 (3.37%) patients, and Grade 3 in 1 (0.67%) patient which was statistically significant (P < 0.001). There was no case of Grade 4-impossible mask ventilation (IMV) in either group [Table 3] and [Figure 2].
Difficult laryngoscopy
Difficulty in laryngoscopy was evaluated using Cormack–Lehane grading. Higher CL grades were observed in Group H in comparison to Group L. There was no case of CL grade 4 [Table 4].
Difficult intubation
A failed intubation attempt is defined as an attempt where the trachea was not intubated with three attempts.The hypoxic event was defined as a decrease in oxygen saturation (SpO2 <90%) during attempts at securing airway management by an unassisted anesthesiologist.
A higher incidence of difficult intubation was observed in Group H (56.9% vs. 11.5%, P < 0.001). In Group H, 24 (23.52%) patients had easy IDS scores, 32 (31.37%) had slight difficulty, and 46 (45.09%) had moderate-to-major difficulty. In Group L, 131 (88.5%) patients had easy IDS scores, 15 (10.13%) had slight difficulty, and 2 (1.35%) had moderate-to-major difficulty. This was statistically significant (P < 0.000077); [Table 5]. The receiver operating characteristic curve for IDS score shows area under the ROC curve of 93.1% as shown in [Figure 3].
Figure 3: ROC curve - IDS score. ROC = Receiver operating characteristic, IDS = Intubation difficult scoreSecondary outcomes
Airway complications
Airway trauma includes blood on lips, teeth, oral mucosa, or on the device while intubation. Hemodynamic complications include laryngospasm, bronchospasm, vocal cord paralysis, arytenoid injury, and tracheal or esophageal perforation. Incidence of bleeding, injury to teeth or mucosa was higher in Group H than in Group L (P < 0.0001) [Table 6].
Intraoperative hemodynamic complications were seen more frequently in Group H than Group L. Hypoxia was observed in 12 patients in Group H, whereas it was not observed in Group L [Table 7].
Discussion The main finding of this study was that STOP-BANG questionnaire is a reliable, easy, quick, and cost-effective tool in identifying patients at risk of difficult airway preoperatively.
Undiagnosed OSA poses a great risk in patients undergoing elective surgeries under general anesthesia as they have a greater incidence of DMV or difficult intubation (DI) or IMV.[6],[7] Even asymptomatic OSA is independently associated with an increased morbidity and mortality. Moderate OSA causes hypertension and neurocognitive impairment, while severe OSA and OSAS predispose to coronary artery disease, stroke, and increased mortality. A prospective cohort study[8] showed that untreated OSA was independently associated with more cardiopulmonary complications, unplanned re-intubations, and myocardial infarction. Untreated OSA by itself is an independent risk factor for cardiovascular diseases.[9],[10] However, Chung et al. found that 82% of men and 92% of women with moderate-to-severe sleep apnea have not been diagnosed.[11] Finkel et al.[12] showed 23% positive screening of OSA among the surgical population in an academic center.
Chung et al. also found that a STOP-BANG score ≥3 has a high sensitivity of 90% and a high positive predictive value of 85% in diagnosing a patient with OSA.[13] Other studies[14],[15] also use STOP-BANG >3 as cutoff for OSA owing to high specificity and positive predictive value. Hence, in our study, we have used a score of ≥3 as high risk for OSA.
Demographic parameters such as age and BMI are higher in patients scoring 3 or more on STOP-BANG questionnaire which is similar to the studies were done by Mathangi et al.,[14] Agrawal et al.,[16] and Magalhães et al.[17] The severity of OSA was not found to correlate with BMI in few population studies done on obese patients by Kang et al.[18] and Chung et al.[13] This could have been because the severity of OSA depends on multiple factors apart from BMI such as upper airway collapsibility, pharyngeal dilator tone, arousal response in OSA, and pharyngeal neuropathy. In our study, we observed that BMI >35 kg/m2 had a positive predictive value of 97.61% for difficult airway (confidence interval [CI]: 64.93%–74.6%).
Significantly, airway parameters including interincisor gap and sternomental and thyromental distance were observed to be lesser in Group H than in Group L; Mallampati grade and neck circumference were found to be higher in Group H. We observed that neck circumference >40 cm had a positive predictive value of 94.4% for difficult airway (CI: 68.9%–78.9%). A history of loud snoring had an accuracy of 85.6% for identifying difficult airway.
Mathangi et al.[14] demonstrated 51% incidence of DMV in patients fulfilling STOP-BANG criteria. Magalhães et al.,[17] in their study, observed 16.3% incidence of DMV in the obese group. Corso et al.[19] observed 23% incidence of DMV in high-risk OSA group (95% CI: 19%–27%). In this study, the proportion of patients with DMV during induction was 59.8% in Group H versus 4.05% in Group L. No patient had IMV.
We also observed a significantly higher incidence of difficult intubation in Group H (56.9% vs. 11.5%, P < 0.001). Our study findings are comparable to those of Mathangi et al.,[14] in which DIT incidence was significantly higher in the OSA group as compared to non-OSA group (22.2% vs. 4.3%). There was 20% incidence of higher CL grades as observed by Kim and Lee.[20]
Corso et al.[19] observed 20% incidence of DIT in high-risk OSA group (95% CI: 17%–24%).
Pera et al.[21] found a higher frequency of Cormack Grade III/IV in high-risk group (77.8% vs. 2.6%; P < 0.0001). The incidence of DIT was 21.9% among OSA patients in a study by Siyam and Benhamou.[22]
Airway complications were more often encountered in Group H owing to repeated manipulation of the airway or change of technique or operators. Airway complications were not evaluated in detail in other studies so far. Hypoxia during mask ventilation was observed in 12 patients in Group H. None of the patients in the present study were allowed to experience desaturation below 85% as immediate interventions such as adding airway adjuncts or employment of two-person technique for mask ventilation were practiced. Hypoxic events were not observed in Group L. One patient in Group L had an incidence of laryngospasm immediately after extubation which was promptly treated. Corso et al.[19] observed higher incidence of postoperative complications in patients with high risk for OSA 9% (95% CI: 6%–11%) which is similar to our study. Chia et al.,[23] Vasu et al.,[2] and Chung et al.[11] found that a higher STOP-BANG score was associated with an increased rate of postoperative cardiac and pulmonary complications.
The main limitation in our study is that a larger sample size would have been more adequate to confirm our results. Second, mask ventilation was not reassessed after the administration of muscle relaxant as few studies suggest that the degree of difficulty becomes lesser following muscle relaxation.
Conclusion STOP-BANG questionnaire is a reliable, cost-effective bedside screening tool which is a significant predictor of difficult airway owing to undiagnosed OSA in general surgical patients. It helps the anesthesiologist to anticipate and manage difficult airway, thereby decreasing the rate of perioperative complications. Hence, the inclusion of STOP-BANG criteria in routine preanesthetic evaluation is suggested for safe conduct of anesthesia. Further studies to evaluate the utility of STOP-BANG questionnaire to correlate with newer airway parameters should be considered.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References 
Comments (0)