International Journal of Pediatric Otorhinolaryngology 77 (2013) 2044–2048
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Association of childhood high body mass index and sleep disordered breathing with perioperative laryngospasm Olubukola O. Nafiu *, Yosha Prasad, Wilson T. Chimbira Department of Anesthesiology, Section of Pediatric Anesthesiology, University of Michigan, Ann Arbor, MI, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 17 July 2013 Received in revised form 30 September 2013 Accepted 3 October 2013 Available online 14 October 2013
Background: Childhood high body mass index (BMI) and sleep disordered breathing (SDB) are increasingly prevalent and both are associated with perioperative respiratory complications. Laryngospasm is one of the more serious respiratory complications with potentially devastating consequences. It is presently unclear whether high BMI and incident SDB in children significantly amplifies the risks of perioperative laryngospasm. This study examined the hypothesis that compared to controls; children with high BMI and SDB at the time of surgery have higher rates of perioperative laryngospasm. Methods: Children (6–18 yr) who underwent elective, non-cardiac operations at a tertiary care center were the subjects of this cross-sectional study. Rates of perioperative laryngospasm were compared between normal controls and children who were overweight/obese and had clinical history of SDB at the time of surgery. Stepwise logistic regression analysis was performed to identify independent predictors of perioperative laryngospasm (dependent variable) using high BMI/SDB as the primary predictor variable. Results: Among 642 children, those who were overweight/obese and had incident SDB (N = 197) were younger, and had higher indexes of central adiposity. Children with high BMI and SDB had 3.8 times higher unadjusted odds of developing intraoperative laryngospasm (OR = 3.8; 95% CI = 2.1–6.9, p < 0.001). After adjusting for several relevant covariates, the following factors were found to be independent predictors of perioperative laryngospasm: high BMI + SDB, male sex and increasing neck circumference. Conclusion: High BMI and incident SDB in children is associated with increased rates of perioperative laryngospasm. The mechanism(s) underlying this propensity to laryngospasm deserve further elucidation. Published by Elsevier Ireland Ltd.
Keywords: Childhood obesity Sleep disordered breathing Laryngospasm General anesthesia Risk factors
1. Introduction Childhood high body mass index (BMI) has become one of the foremost issues in contemporary bio-medical research particularly because of its importance in predicting overweight and obesity in adulthood as well as its association with various cardiovascular risk factors [1,2]. Recent epidemiologic trends also indicate a steady increase in the prevalence of sleep disordered breathing (SDB) in children and adolescents, particularly among those with high BMI [3,4]. SDB is a spectrum of disorders encompassing habitual snoring, obstructive hypoventilation (OH), and varying degrees of obstructive sleep apnea (OSA) [4,5]. Frequently, high BMI and SDB overlap or may occur concurrently in the same
* Corresponding author at: 1500 East Medical Centre Drive, University of Michigan Health System, Room UH 1H247, Ann Arbor, MI 48109-0048, USA. Tel.: +1 734 936 4280; fax: +1 734 936 9091. E-mail address: onafi[email protected] (O.O. Nafiu). 0165-5876/$ – see front matter . Published by Elsevier Ireland Ltd. http://dx.doi.org/10.1016/j.ijporl.2013.10.004
patient and may indeed potentiate the adverse impact of each other [2,6]. Because these ‘‘twin protagonists’’ are so pervasive in the general population, it becomes increasingly likely that children undergoing anesthesia and surgery will have one or both disorders with the implicit potential to increase the risk of perioperative complications, particularly, perioperative airway complications. Laryngospasm is a serious perioperative adverse event with potentially devastating consequences [7]. About one in three children who develop laryngospasm suffer significant physiological perturbation [8,9]. Indeed, occurrence of laryngospasm in children is frequently associated with rapid escalation of care in the perioperative period and could be a major source of stress for the anesthesia caregiver, surgeon and operating room personnel. Several risk factors for perioperative laryngospasm have been described [10–12]. Among these, airway irritation with or without inflammation is commonly cited as a potential risk factor for laryngospasm [13]. Although high BMI and SDB are frequently cited as risk factors for perioperative airway complications [14,15], it is presently unknown whether children who are overweight or
O.O. Nafiu et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 2044–2048
obese and have a concurrent diagnosis of SDB have increased rates of perioperative laryngospasm. Therefore, the primary objective of this investigation was to determine whether, compared to normal controls, rates of perioperative laryngospasm are increased in children with high BMI and SDB. We tested the hypothesis that children who were overweight/obese and have incident diagnosis of SDB at the time of surgery would have higher rates of perioperative laryngospasm. 2. Methods Data for the present report were derived from a prospective observational study, which explored the association of clothing size with adiposity indices in children who underwent elective, non-cardiac operations [16]. The study was approved by our Institutional Review Board and written parental or caregiver consent and patient assent (where applicable) were obtained. 2.1. Clinical, anthropometric measurements and operational definition of terms Trained research assistants (RAs) who were unaware of the study’s hypothesis collected all clinical and anthropometric data. Age in years, sex, American Society of Anesthesiology (ASA) classification and ethnicity were recorded in all study subjects. Height was measured to the nearest 0.1 cm using a wall-mounted stadiometer with the patients shoeless and head held in the horizontal plane. Body weight was measured, to the nearest 0.1 kg using a calibrated electronic weighing scale with patients lightly clad in hospital gowns. Neck circumference (NC) was measured with a flexible tape, with study subjects in the standing position, head held erect, at the level of the thyroid cartilage. Waist circumference (WC) was measured (to the nearest 0.1 cm), with the children standing, at the end of normal expiration, using a flexible tape at a point midway between the inferior margin of the lowest rib and the iliac crest. Measurements were taken with the tape snug but not compressing the skin. In addition to the primary surgical diagnoses, the presence of secondary diagnoses like diabetes (type I or II), bronchial asthma and hypertension was noted. Diagnosis of bronchial asthma was based on history of use of bronchodilators or on a physician diagnosis. Clinical diagnosis of SDB was based on parental or caregiver response to the question: ‘‘have you ever been told that your child has OSA?’’ Subgroup assignment was based on the response to this question and on the presence of habitual snoring (defined as parental or caregiver report of loud snoring in the child for at least 3 or more nights per week) [17]. As an index of central adiposity, abdominal obesity was defined as age and genderadjusted WC > 90th percentile [18]. Post anesthesia care unit (PACU) airway obstruction indicates airflow obstruction requiring jaw/chin manipulation or use of oral/nasal airway to restore adequate air movement. Difficult mask ventilation (DMV) was graded according to Han et al. [19]. Diagnosis of laryngospasm was entirely clinical and based on its report as a quality assurance (QA) adverse event in the patient’s chart. Only cases requiring continuous positive airway pressure (CPAP) or low dose succinylcholine are reportable as laryngospasm. In addition, upon arrival in the PACU, anesthesia providers were specifically asked whether any of the study patients had laryngospasm during the perioperative period. 2.2. Statistical analysis Data analyses were performed with Statistical Package for the Social Sciences (SPSS for Windows, Chicago, IL, USA, version 19.0). Means and standard deviations of demographic and
2045
anthropometric variables were compared along gender lines. Continuous variables (age, height, weight and BMI) were examined for normal distribution with the Kolmogorov– Smirnov test. We calculated BMI as weight in kilograms divided by the square of the height in meters (BMI = kg/m2) for all patients. We then transformed BMI into a categorical variable for the grouping of children into two categories thus, normal or overweight/obese. Normal BMI indicates age and sex-specific BMI between the 5th and 84th percentile, while high BMI (overweight/obese) indicates age and sex-specific BMI 85th percentile according to reference growth charts from the National Center for Health Statistics (NCHS)/Centers for Disease Control and Prevention (CDC) [20]. Further stratification based on preoperative history of SDB was done to yield two groups: high BMI/SDB and normal BMI/non-SDB. Rates of perioperative laryngospasm were compared between the two groups. We excluded (83) children with either high BMI but no SDB or normal BMI and positive SDB from univariate and multivariate group comparisons. Baseline differences between the high BMI/SDB group and controls were explored using Pearson’s Chi-square (for categorical variables) and one-way ANOVA (for continuous variables). Incidence of laryngospasm was described as simple proportions and its rates were compared between cases and controls. Univariate differences between cases (laryngospasm) and controls were examined, and odds ratios with corresponding confidence intervals calculated as necessary. 2.2.1. Multivariate analysis The main objective of this study was to explore the association between high BMI/SDB and rates of perioperative laryngospasm (the exposure variable). Variables used in the multivariate analysis were selected on the basis of statistical significance (differences between the exposure and predictor variables) and clinical relevance and included age, gender, neck circumference, high BMI/SDB, inhalational vs. intravenous induction and presence of supraglottic or endotracheal tube. Prior to performing multiple logistic regression analyses, we examined the variables for multi-colinearity by first creating a correlation matrix and testing for highly correlated variables (0.7). After co-linearity was assessed, WC was found to have a higher correlation with BMI than NC (0.86 vs. 0.69). Therefore, NC was included in the regression model while WC was left out. We also calculated the variance inflation factor (VIF) for a model including the three measures of adiposity and NC had a moderately low value of 2.66. Nevertheless, we ran a separate regression model including WC (result not shown). All selected variables were entered into a non-parsimonious logistic regression model to determine independent predictors of perioperative laryngospasm. All variables with a p value 0.05 were considered independent predictors of laryngospasm. Goodness of fit of the logistic regression model was assessed with the Hosmer–Lemeshow test [21]. Furthermore, the model’s predictive value was assessed using Receiver operating characteristic (ROC) analyses and area under the curve (AUC). 3. Results 3.1. Study cohort Our study population comprised of 725 children (6–17 yr) of whom 52.3% were boys. The overall prevalence of high BMI (overweight/obese) was 31.4% while the overall prevalence of SDB was 15.2%. A total of 197 (27.2%) children were classifiable as having high BMI/SDB. After excluding children with either high BMI or SDB (N = 83), data on the remaining 642 children were utilized for the remaining analyses. In this cohort, laryngospasm occurred in 48 (7.4%) patients.
2046
O.O. Nafiu et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 2044–2048
Table 1 Baseline characteristics of the study subjects by BMI and SDB status. High BMI + SDB (N = 197) Continuous variables Age (years) Height (cm) Weight (kg) BMI (kg/m2) NC (cm) WC (cm) SBP (mmHg) DBP (mmHg) Categorical variables (%) Asthma Abdominal obesity Inhalational induction Airway maintained with ETT Difficult mask ventilation OR laryngospasm PACU airway obstruction
9.7 3.7 142.3 19.3 56.7 29.1 26.5 9.2 34.6 5.8 86.0 20.4 113.7 14.7 65.7 9.1 34.5 50.8 65.5 69.5 15.5 13.6 7.1
Normal controls (N = 445) 11.1 3.7 146.2 21.8 39.6 16.3 17.5 2.7 30.3 3.9 65.7 10.3 110.9 13.9 64.7 9.3 16.9 13.6 63.4 56.6 2.0 4.0 2.7
p value
0.007 0.029
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Association of childhood high body mass index and sleep disordered breathing with perioperative laryngospasm Olubukola O. Nafiu *, Yosha Prasad, Wilson T. Chimbira Department of Anesthesiology, Section of Pediatric Anesthesiology, University of Michigan, Ann Arbor, MI, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 17 July 2013 Received in revised form 30 September 2013 Accepted 3 October 2013 Available online 14 October 2013
Background: Childhood high body mass index (BMI) and sleep disordered breathing (SDB) are increasingly prevalent and both are associated with perioperative respiratory complications. Laryngospasm is one of the more serious respiratory complications with potentially devastating consequences. It is presently unclear whether high BMI and incident SDB in children significantly amplifies the risks of perioperative laryngospasm. This study examined the hypothesis that compared to controls; children with high BMI and SDB at the time of surgery have higher rates of perioperative laryngospasm. Methods: Children (6–18 yr) who underwent elective, non-cardiac operations at a tertiary care center were the subjects of this cross-sectional study. Rates of perioperative laryngospasm were compared between normal controls and children who were overweight/obese and had clinical history of SDB at the time of surgery. Stepwise logistic regression analysis was performed to identify independent predictors of perioperative laryngospasm (dependent variable) using high BMI/SDB as the primary predictor variable. Results: Among 642 children, those who were overweight/obese and had incident SDB (N = 197) were younger, and had higher indexes of central adiposity. Children with high BMI and SDB had 3.8 times higher unadjusted odds of developing intraoperative laryngospasm (OR = 3.8; 95% CI = 2.1–6.9, p < 0.001). After adjusting for several relevant covariates, the following factors were found to be independent predictors of perioperative laryngospasm: high BMI + SDB, male sex and increasing neck circumference. Conclusion: High BMI and incident SDB in children is associated with increased rates of perioperative laryngospasm. The mechanism(s) underlying this propensity to laryngospasm deserve further elucidation. Published by Elsevier Ireland Ltd.
Keywords: Childhood obesity Sleep disordered breathing Laryngospasm General anesthesia Risk factors
1. Introduction Childhood high body mass index (BMI) has become one of the foremost issues in contemporary bio-medical research particularly because of its importance in predicting overweight and obesity in adulthood as well as its association with various cardiovascular risk factors [1,2]. Recent epidemiologic trends also indicate a steady increase in the prevalence of sleep disordered breathing (SDB) in children and adolescents, particularly among those with high BMI [3,4]. SDB is a spectrum of disorders encompassing habitual snoring, obstructive hypoventilation (OH), and varying degrees of obstructive sleep apnea (OSA) [4,5]. Frequently, high BMI and SDB overlap or may occur concurrently in the same
* Corresponding author at: 1500 East Medical Centre Drive, University of Michigan Health System, Room UH 1H247, Ann Arbor, MI 48109-0048, USA. Tel.: +1 734 936 4280; fax: +1 734 936 9091. E-mail address: onafi[email protected] (O.O. Nafiu). 0165-5876/$ – see front matter . Published by Elsevier Ireland Ltd. http://dx.doi.org/10.1016/j.ijporl.2013.10.004
patient and may indeed potentiate the adverse impact of each other [2,6]. Because these ‘‘twin protagonists’’ are so pervasive in the general population, it becomes increasingly likely that children undergoing anesthesia and surgery will have one or both disorders with the implicit potential to increase the risk of perioperative complications, particularly, perioperative airway complications. Laryngospasm is a serious perioperative adverse event with potentially devastating consequences [7]. About one in three children who develop laryngospasm suffer significant physiological perturbation [8,9]. Indeed, occurrence of laryngospasm in children is frequently associated with rapid escalation of care in the perioperative period and could be a major source of stress for the anesthesia caregiver, surgeon and operating room personnel. Several risk factors for perioperative laryngospasm have been described [10–12]. Among these, airway irritation with or without inflammation is commonly cited as a potential risk factor for laryngospasm [13]. Although high BMI and SDB are frequently cited as risk factors for perioperative airway complications [14,15], it is presently unknown whether children who are overweight or
O.O. Nafiu et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 2044–2048
obese and have a concurrent diagnosis of SDB have increased rates of perioperative laryngospasm. Therefore, the primary objective of this investigation was to determine whether, compared to normal controls, rates of perioperative laryngospasm are increased in children with high BMI and SDB. We tested the hypothesis that children who were overweight/obese and have incident diagnosis of SDB at the time of surgery would have higher rates of perioperative laryngospasm. 2. Methods Data for the present report were derived from a prospective observational study, which explored the association of clothing size with adiposity indices in children who underwent elective, non-cardiac operations [16]. The study was approved by our Institutional Review Board and written parental or caregiver consent and patient assent (where applicable) were obtained. 2.1. Clinical, anthropometric measurements and operational definition of terms Trained research assistants (RAs) who were unaware of the study’s hypothesis collected all clinical and anthropometric data. Age in years, sex, American Society of Anesthesiology (ASA) classification and ethnicity were recorded in all study subjects. Height was measured to the nearest 0.1 cm using a wall-mounted stadiometer with the patients shoeless and head held in the horizontal plane. Body weight was measured, to the nearest 0.1 kg using a calibrated electronic weighing scale with patients lightly clad in hospital gowns. Neck circumference (NC) was measured with a flexible tape, with study subjects in the standing position, head held erect, at the level of the thyroid cartilage. Waist circumference (WC) was measured (to the nearest 0.1 cm), with the children standing, at the end of normal expiration, using a flexible tape at a point midway between the inferior margin of the lowest rib and the iliac crest. Measurements were taken with the tape snug but not compressing the skin. In addition to the primary surgical diagnoses, the presence of secondary diagnoses like diabetes (type I or II), bronchial asthma and hypertension was noted. Diagnosis of bronchial asthma was based on history of use of bronchodilators or on a physician diagnosis. Clinical diagnosis of SDB was based on parental or caregiver response to the question: ‘‘have you ever been told that your child has OSA?’’ Subgroup assignment was based on the response to this question and on the presence of habitual snoring (defined as parental or caregiver report of loud snoring in the child for at least 3 or more nights per week) [17]. As an index of central adiposity, abdominal obesity was defined as age and genderadjusted WC > 90th percentile [18]. Post anesthesia care unit (PACU) airway obstruction indicates airflow obstruction requiring jaw/chin manipulation or use of oral/nasal airway to restore adequate air movement. Difficult mask ventilation (DMV) was graded according to Han et al. [19]. Diagnosis of laryngospasm was entirely clinical and based on its report as a quality assurance (QA) adverse event in the patient’s chart. Only cases requiring continuous positive airway pressure (CPAP) or low dose succinylcholine are reportable as laryngospasm. In addition, upon arrival in the PACU, anesthesia providers were specifically asked whether any of the study patients had laryngospasm during the perioperative period. 2.2. Statistical analysis Data analyses were performed with Statistical Package for the Social Sciences (SPSS for Windows, Chicago, IL, USA, version 19.0). Means and standard deviations of demographic and
2045
anthropometric variables were compared along gender lines. Continuous variables (age, height, weight and BMI) were examined for normal distribution with the Kolmogorov– Smirnov test. We calculated BMI as weight in kilograms divided by the square of the height in meters (BMI = kg/m2) for all patients. We then transformed BMI into a categorical variable for the grouping of children into two categories thus, normal or overweight/obese. Normal BMI indicates age and sex-specific BMI between the 5th and 84th percentile, while high BMI (overweight/obese) indicates age and sex-specific BMI 85th percentile according to reference growth charts from the National Center for Health Statistics (NCHS)/Centers for Disease Control and Prevention (CDC) [20]. Further stratification based on preoperative history of SDB was done to yield two groups: high BMI/SDB and normal BMI/non-SDB. Rates of perioperative laryngospasm were compared between the two groups. We excluded (83) children with either high BMI but no SDB or normal BMI and positive SDB from univariate and multivariate group comparisons. Baseline differences between the high BMI/SDB group and controls were explored using Pearson’s Chi-square (for categorical variables) and one-way ANOVA (for continuous variables). Incidence of laryngospasm was described as simple proportions and its rates were compared between cases and controls. Univariate differences between cases (laryngospasm) and controls were examined, and odds ratios with corresponding confidence intervals calculated as necessary. 2.2.1. Multivariate analysis The main objective of this study was to explore the association between high BMI/SDB and rates of perioperative laryngospasm (the exposure variable). Variables used in the multivariate analysis were selected on the basis of statistical significance (differences between the exposure and predictor variables) and clinical relevance and included age, gender, neck circumference, high BMI/SDB, inhalational vs. intravenous induction and presence of supraglottic or endotracheal tube. Prior to performing multiple logistic regression analyses, we examined the variables for multi-colinearity by first creating a correlation matrix and testing for highly correlated variables (0.7). After co-linearity was assessed, WC was found to have a higher correlation with BMI than NC (0.86 vs. 0.69). Therefore, NC was included in the regression model while WC was left out. We also calculated the variance inflation factor (VIF) for a model including the three measures of adiposity and NC had a moderately low value of 2.66. Nevertheless, we ran a separate regression model including WC (result not shown). All selected variables were entered into a non-parsimonious logistic regression model to determine independent predictors of perioperative laryngospasm. All variables with a p value 0.05 were considered independent predictors of laryngospasm. Goodness of fit of the logistic regression model was assessed with the Hosmer–Lemeshow test [21]. Furthermore, the model’s predictive value was assessed using Receiver operating characteristic (ROC) analyses and area under the curve (AUC). 3. Results 3.1. Study cohort Our study population comprised of 725 children (6–17 yr) of whom 52.3% were boys. The overall prevalence of high BMI (overweight/obese) was 31.4% while the overall prevalence of SDB was 15.2%. A total of 197 (27.2%) children were classifiable as having high BMI/SDB. After excluding children with either high BMI or SDB (N = 83), data on the remaining 642 children were utilized for the remaining analyses. In this cohort, laryngospasm occurred in 48 (7.4%) patients.
2046
O.O. Nafiu et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 2044–2048
Table 1 Baseline characteristics of the study subjects by BMI and SDB status. High BMI + SDB (N = 197) Continuous variables Age (years) Height (cm) Weight (kg) BMI (kg/m2) NC (cm) WC (cm) SBP (mmHg) DBP (mmHg) Categorical variables (%) Asthma Abdominal obesity Inhalational induction Airway maintained with ETT Difficult mask ventilation OR laryngospasm PACU airway obstruction
9.7 3.7 142.3 19.3 56.7 29.1 26.5 9.2 34.6 5.8 86.0 20.4 113.7 14.7 65.7 9.1 34.5 50.8 65.5 69.5 15.5 13.6 7.1
Normal controls (N = 445) 11.1 3.7 146.2 21.8 39.6 16.3 17.5 2.7 30.3 3.9 65.7 10.3 110.9 13.9 64.7 9.3 16.9 13.6 63.4 56.6 2.0 4.0 2.7
p value
0.007 0.029
