Surg Endosc DOI 10.1007/s00464-013-3295-7
and Other Interventional Techniques
Predicting sleep apnea in morbidly obese adolescents undergoing bariatric surgery Emily S. Koeck • Leah C. Barefoot • Miller Hamrick • Judith A. Owens • Faisal G. Qureshi • Evan P. Nadler
Received: 6 June 2013 / Accepted: 18 October 2013 Ó Springer Science+Business Media New York 2013
Abstract Introduction In adults, the association between obesity and obstructive sleep apnea (OSA) is established, and many are concerned OSA increases surgical risk. Pre-operative screening for OSA is standard, and this is also the case in adolescent bariatric surgery. We noted many of our patients were without significant OSA, despite being obese. We reviewed our experience with screening polysomnography (PSG) to determine any predictive variables or complications. Methods All bariatric surgery patients from our hospital who had undergone PSG were included, and were stratified into ‘OSA’ or ‘no OSA’ by obstructive apnea-hypopnea index (OAHI), as well as by sex. Results A total of 49 adolescents enrolled during the study period: 10 males and 39 females. OSA prevalence was 42.9 %; males 80 %, females 33.3 %. Height, weight, body mass index (BMI), and prevalence of hypertension were significantly higher in patients with OSA. By sex, females also had more metabolic syndrome and witnessed
Research presented in part at 2013 Academic Surgical Congress, New Orleans, LA, February 5–7, 2013 E. S. Koeck (&) L. C. Barefoot M. Hamrick F. G. Qureshi E. P. Nadler Division of Pediatric Surgery, Children’s National Medical Center, 111 Michigan Ave NW, Washington, DC 20010, USA e-mail: [email protected] E. S. Koeck E. P. Nadler Sheikh Zayed Institute for Pediatric Surgical Innovation, Children’s National Medical Center, 111 Michigan Ave NW, Washington, DC 20010, USA J. A. Owens Division of Sleep Medicine, Children’s National Medical Center, Washington, DC 20010, USA
apneas, while only weight and BMI remained significant in males. There were no peri-operative complications. Conclusions Despite uniform obesity, less than half our adolescents had significant OSA on PSG. As no modeling exists to predict OSA in morbidly obese adolescents, we continue to recommend routine PSG, especially in higher weight and BMI patients, and those with hypertension. Keywords Obesity Bariatric surgery Obstructive sleep apnea Pediatric surgery
Introduction The obesity epidemic now includes a new generation of morbidly obese children and adolescents, for whom weightloss surgery is being more commonly recommended. Failure of traditional medical and behavioral weight management programs is the primary indication for surgical intervention [1, 2]. Many have questioned whether weight-loss surgery is appropriate for this patient population, and pediatricians often do not refer adolescents for surgery. In regional and national surveys, nearly half of all pediatricians would never refer an adolescent for bariatric surgery, often citing the risks of the procedure as a primary reason for their reluctance [3, 4]. If there were substantial convincing data that weight-loss surgery in adolescents is indeed safe, perhaps the reluctance of pediatricians to refer for bariatric surgery could be abrogated. Obstructive sleep apnea (OSA) is one of the many obesity-related conditions that negatively affects adolescents with morbid obesity and is of special concern due to not only its impact on the patient if it is left untreated, but also to its potential bearing on peri-operative outcomes. In the general population, untreated OSA carries an increased risk of all-cause morbidity and mortality [5–7];
123
Surg Endosc
emerging evidence suggests these risks also affect children and adolescents with OSA [8]. It is assumed that these risks translate into an increased risk of peri-operative complications associated with general anesthesia [5, 9–11]. Altered respiratory physiology and mechanics predispose patients with OSA to intra-operative hypoxemia and hypercapnia, increasing the risk of arrhythmia and hypertension [12, 13]. Due to altered drug metabolism, obese patients and especially those with OSA, require careful selection of sedatives, anesthetics, and analgesics [12, 14– 17]. Further complicating the care of the morbidly obese patient with OSA is the lack of evidence-based guidelines to guide the use of anesthetics in these patients [17]. Available data in adolescent cohorts are particularly limited in this regard, and the incidence of OSA in adolescents with obesity and its impact on anesthetic outcome has been far ranging and not fully elucidated [1, 18–20]. Thus, we evaluated our experience with OSA in our cohort of adolescents with morbid obesity presenting for bariatric surgery to determine the prevalence of disease and impact on surgical outcomes, and whether we could predict which patients were at highest risk for having the disease. Specifically, we reviewed our experience with polysomnography (PSG) to identify any potential predictors of significant OSA, as well as to determine if a pre-operative diagnosis of significant OSA was associated with increased incidence of peri-operative complications.
Methods We conducted a retrospective review of the first 50 adolescents with morbid obesity who underwent bariatric surgery (laparoscopic sleeve gastrectomy or laparoscopic gastric band) at our institution from January 2010 until April 2013. All patients who successfully completed surgery were eligible for inclusion. All patients underwent standard PSG as part of their pre-surgical work up, the results of which were interpreted by hospital sleep physicians. The Obstructive apnea-hypopnea index (OAHI) was calculated and defined as the number of obstructive apneas and hypopneas occurring per hour of sleep. The presence of OSA was determined when OAHI was C5, in accordance with standard diagnostic guidelines [21]. We also evaluated our cohort using a stricter diagnostic cut-off of OAHI C10, to determine any differences in patients with severe OSA. Demographic variables, including age, sex, ethnicity, height, weight, body mass index (BMI), and presence of sleep apnea symptoms were collected at the time of PSG. Sleep apnea symptoms evaluated at time of PSG included somnolence, snoring, morning headache, restless sleep, and witnessed apneas. A chart review was later conducted to
123
determine comorbid conditions, including hypertension, type 2 diabetes or insulin resistance, metabolic syndrome, asthma, or gastroesophageal reflux disease. History of adenotonsillectomy and the incidence of peri-operative complications were also determined from chart review. Demographic and clinical variables were compared between patients with OSA and those without OSA. Categorical variables were compared using Fisher’s exact test, while continuous variables were compared using Student’s t test.
Results A total of 54 adolescents underwent bariatric surgery at our institution during the study period. Of those, 49 were enrolled in the study; 10 males and 39 females. Due to relatively low BMI (36) and lack of OSA symptoms, one patient did not have a sleep study prior to surgery. Three patients who were admitted prior to surgery to an inpatient weight loss facility that did not have the capability to perform a formal PSG were also excluded. One patient with Down syndrome was excluded, due to the physiologic differences and higher incidence of OSA in patients with Down syndrome. The rest of the cohort included 29 African-Americans, nine Hispanics, ten Caucasians, and one patient of mixed Caucasian and AfricanAmerican descent. Mean age was 16.2 ± 2.1 years. Mean weight was 144.3 ± 32.4 kg, with mean BMI of 49.9 ± 8.8. Full demographic characteristics are detailed in Table 1. Male patients were significantly taller than female patients (179.0 ± 5.5 vs. 166.8 ± 9.2 cm; p \ 0.001) and more likely to have had adenotonsillectomy (70.0 vs 28.2 %; p = 0.025), but there were no other demographic differences between sexes. When using OAHI C5 as the diagnostic cut-off, 21 of 49 patients (42.9 %) had OSA. Males had a significantly higher prevalence of OSA (80.0 vs. 33.3 %; p = 0.01). Height (174.0 ± 9.3 vs. 165.7 ± 8.9 cm; p = 0.003), weight (160.9 ± 30.3 vs. 131.8 ± 28.5 kg; p = 0.001), and BMI (53.0 ± 9.2 vs 47.6 ± 7.9; p = 0.04) were significantly higher in patients with OSA. Hypertension was significantly more common in patients with OSA (52.3 vs. 14.3 %; p = 0.006). There were no significant associations between OSA and age, comorbidities other than hypertension, or presence of OSA symptoms. At the stricter cut-off of OAHI C10, height, weight, BMI, and presence of hypertension remained significant. However, presence of somnolence became significant, with fewer patients with OSA reporting daytime somnolence (35.2 vs. 75.0 %; p = 0.013). See Tables 2 and 3 for full details of cohort variables. There were no anesthetic- or OSA-related complications in the study cohort, and no patients required new continuous positiveairway pressure (CPAP) to treat OSA during their surgical
Surg Endosc Table 1 Cohort demographics
Cohort
Male
Female
49 (100.0)
10 (20.4)
39 (79.6)
Caucasian
10 (20.4)
2 (20.0)
8 (20.5)
Hispanic
9 (18.4)
3 (30.0)
6 (15.4)
African American
29 (59.2)
5 (50.0)
24 (61.5)
Other
1 (2.0)
0 (0.0)
1 (2.6)
Hypertension
15 (30.6)
5 (50.0)
10 (25.6)
DM/IR
30 (61.2)
8 (80.0)
22 (56.4)
0.28
Metabolic syndrome
6 (12.2)
1 (10.0)
5 (12.8)
1
GERD
9 (18.4)
0 (0.0)
9 (23.1)
0.17
Asthma
25 (51.0)
5 (50.0)
20 (51.3)
1
Somnolence
30 (61.2)
5 (50.0)
25 (64.1)
0.48
Snoring Headache
40 (81.6) 4 (8.2)
10 (100.0) 2 (20.0)
30 (76.9) 2 (5.1)
0.17 0.18
Restless sleep
12 (24.5)
2 (20.0)
10 (25.6)
1
Witnessed apnea
9 (18.4)
2 (20.0)
7 (17.9)
1
History of adenotonsillectomy
18 (36.7)
7 (70.0)
11 (28.2)
0.025
Age mean (years)
16.2 ± 2.1
16.9 ± 2.3
16.0 ± 2.0
0.29
Height mean (cm)
169.3 – 9.9
179.0 – 5.5
166.8 – 9.2
and Other Interventional Techniques
Predicting sleep apnea in morbidly obese adolescents undergoing bariatric surgery Emily S. Koeck • Leah C. Barefoot • Miller Hamrick • Judith A. Owens • Faisal G. Qureshi • Evan P. Nadler
Received: 6 June 2013 / Accepted: 18 October 2013 Ó Springer Science+Business Media New York 2013
Abstract Introduction In adults, the association between obesity and obstructive sleep apnea (OSA) is established, and many are concerned OSA increases surgical risk. Pre-operative screening for OSA is standard, and this is also the case in adolescent bariatric surgery. We noted many of our patients were without significant OSA, despite being obese. We reviewed our experience with screening polysomnography (PSG) to determine any predictive variables or complications. Methods All bariatric surgery patients from our hospital who had undergone PSG were included, and were stratified into ‘OSA’ or ‘no OSA’ by obstructive apnea-hypopnea index (OAHI), as well as by sex. Results A total of 49 adolescents enrolled during the study period: 10 males and 39 females. OSA prevalence was 42.9 %; males 80 %, females 33.3 %. Height, weight, body mass index (BMI), and prevalence of hypertension were significantly higher in patients with OSA. By sex, females also had more metabolic syndrome and witnessed
Research presented in part at 2013 Academic Surgical Congress, New Orleans, LA, February 5–7, 2013 E. S. Koeck (&) L. C. Barefoot M. Hamrick F. G. Qureshi E. P. Nadler Division of Pediatric Surgery, Children’s National Medical Center, 111 Michigan Ave NW, Washington, DC 20010, USA e-mail: [email protected] E. S. Koeck E. P. Nadler Sheikh Zayed Institute for Pediatric Surgical Innovation, Children’s National Medical Center, 111 Michigan Ave NW, Washington, DC 20010, USA J. A. Owens Division of Sleep Medicine, Children’s National Medical Center, Washington, DC 20010, USA
apneas, while only weight and BMI remained significant in males. There were no peri-operative complications. Conclusions Despite uniform obesity, less than half our adolescents had significant OSA on PSG. As no modeling exists to predict OSA in morbidly obese adolescents, we continue to recommend routine PSG, especially in higher weight and BMI patients, and those with hypertension. Keywords Obesity Bariatric surgery Obstructive sleep apnea Pediatric surgery
Introduction The obesity epidemic now includes a new generation of morbidly obese children and adolescents, for whom weightloss surgery is being more commonly recommended. Failure of traditional medical and behavioral weight management programs is the primary indication for surgical intervention [1, 2]. Many have questioned whether weight-loss surgery is appropriate for this patient population, and pediatricians often do not refer adolescents for surgery. In regional and national surveys, nearly half of all pediatricians would never refer an adolescent for bariatric surgery, often citing the risks of the procedure as a primary reason for their reluctance [3, 4]. If there were substantial convincing data that weight-loss surgery in adolescents is indeed safe, perhaps the reluctance of pediatricians to refer for bariatric surgery could be abrogated. Obstructive sleep apnea (OSA) is one of the many obesity-related conditions that negatively affects adolescents with morbid obesity and is of special concern due to not only its impact on the patient if it is left untreated, but also to its potential bearing on peri-operative outcomes. In the general population, untreated OSA carries an increased risk of all-cause morbidity and mortality [5–7];
123
Surg Endosc
emerging evidence suggests these risks also affect children and adolescents with OSA [8]. It is assumed that these risks translate into an increased risk of peri-operative complications associated with general anesthesia [5, 9–11]. Altered respiratory physiology and mechanics predispose patients with OSA to intra-operative hypoxemia and hypercapnia, increasing the risk of arrhythmia and hypertension [12, 13]. Due to altered drug metabolism, obese patients and especially those with OSA, require careful selection of sedatives, anesthetics, and analgesics [12, 14– 17]. Further complicating the care of the morbidly obese patient with OSA is the lack of evidence-based guidelines to guide the use of anesthetics in these patients [17]. Available data in adolescent cohorts are particularly limited in this regard, and the incidence of OSA in adolescents with obesity and its impact on anesthetic outcome has been far ranging and not fully elucidated [1, 18–20]. Thus, we evaluated our experience with OSA in our cohort of adolescents with morbid obesity presenting for bariatric surgery to determine the prevalence of disease and impact on surgical outcomes, and whether we could predict which patients were at highest risk for having the disease. Specifically, we reviewed our experience with polysomnography (PSG) to identify any potential predictors of significant OSA, as well as to determine if a pre-operative diagnosis of significant OSA was associated with increased incidence of peri-operative complications.
Methods We conducted a retrospective review of the first 50 adolescents with morbid obesity who underwent bariatric surgery (laparoscopic sleeve gastrectomy or laparoscopic gastric band) at our institution from January 2010 until April 2013. All patients who successfully completed surgery were eligible for inclusion. All patients underwent standard PSG as part of their pre-surgical work up, the results of which were interpreted by hospital sleep physicians. The Obstructive apnea-hypopnea index (OAHI) was calculated and defined as the number of obstructive apneas and hypopneas occurring per hour of sleep. The presence of OSA was determined when OAHI was C5, in accordance with standard diagnostic guidelines [21]. We also evaluated our cohort using a stricter diagnostic cut-off of OAHI C10, to determine any differences in patients with severe OSA. Demographic variables, including age, sex, ethnicity, height, weight, body mass index (BMI), and presence of sleep apnea symptoms were collected at the time of PSG. Sleep apnea symptoms evaluated at time of PSG included somnolence, snoring, morning headache, restless sleep, and witnessed apneas. A chart review was later conducted to
123
determine comorbid conditions, including hypertension, type 2 diabetes or insulin resistance, metabolic syndrome, asthma, or gastroesophageal reflux disease. History of adenotonsillectomy and the incidence of peri-operative complications were also determined from chart review. Demographic and clinical variables were compared between patients with OSA and those without OSA. Categorical variables were compared using Fisher’s exact test, while continuous variables were compared using Student’s t test.
Results A total of 54 adolescents underwent bariatric surgery at our institution during the study period. Of those, 49 were enrolled in the study; 10 males and 39 females. Due to relatively low BMI (36) and lack of OSA symptoms, one patient did not have a sleep study prior to surgery. Three patients who were admitted prior to surgery to an inpatient weight loss facility that did not have the capability to perform a formal PSG were also excluded. One patient with Down syndrome was excluded, due to the physiologic differences and higher incidence of OSA in patients with Down syndrome. The rest of the cohort included 29 African-Americans, nine Hispanics, ten Caucasians, and one patient of mixed Caucasian and AfricanAmerican descent. Mean age was 16.2 ± 2.1 years. Mean weight was 144.3 ± 32.4 kg, with mean BMI of 49.9 ± 8.8. Full demographic characteristics are detailed in Table 1. Male patients were significantly taller than female patients (179.0 ± 5.5 vs. 166.8 ± 9.2 cm; p \ 0.001) and more likely to have had adenotonsillectomy (70.0 vs 28.2 %; p = 0.025), but there were no other demographic differences between sexes. When using OAHI C5 as the diagnostic cut-off, 21 of 49 patients (42.9 %) had OSA. Males had a significantly higher prevalence of OSA (80.0 vs. 33.3 %; p = 0.01). Height (174.0 ± 9.3 vs. 165.7 ± 8.9 cm; p = 0.003), weight (160.9 ± 30.3 vs. 131.8 ± 28.5 kg; p = 0.001), and BMI (53.0 ± 9.2 vs 47.6 ± 7.9; p = 0.04) were significantly higher in patients with OSA. Hypertension was significantly more common in patients with OSA (52.3 vs. 14.3 %; p = 0.006). There were no significant associations between OSA and age, comorbidities other than hypertension, or presence of OSA symptoms. At the stricter cut-off of OAHI C10, height, weight, BMI, and presence of hypertension remained significant. However, presence of somnolence became significant, with fewer patients with OSA reporting daytime somnolence (35.2 vs. 75.0 %; p = 0.013). See Tables 2 and 3 for full details of cohort variables. There were no anesthetic- or OSA-related complications in the study cohort, and no patients required new continuous positiveairway pressure (CPAP) to treat OSA during their surgical
Surg Endosc Table 1 Cohort demographics
Cohort
Male
Female
49 (100.0)
10 (20.4)
39 (79.6)
Caucasian
10 (20.4)
2 (20.0)
8 (20.5)
Hispanic
9 (18.4)
3 (30.0)
6 (15.4)
African American
29 (59.2)
5 (50.0)
24 (61.5)
Other
1 (2.0)
0 (0.0)
1 (2.6)
Hypertension
15 (30.6)
5 (50.0)
10 (25.6)
DM/IR
30 (61.2)
8 (80.0)
22 (56.4)
0.28
Metabolic syndrome
6 (12.2)
1 (10.0)
5 (12.8)
1
GERD
9 (18.4)
0 (0.0)
9 (23.1)
0.17
Asthma
25 (51.0)
5 (50.0)
20 (51.3)
1
Somnolence
30 (61.2)
5 (50.0)
25 (64.1)
0.48
Snoring Headache
40 (81.6) 4 (8.2)
10 (100.0) 2 (20.0)
30 (76.9) 2 (5.1)
0.17 0.18
Restless sleep
12 (24.5)
2 (20.0)
10 (25.6)
1
Witnessed apnea
9 (18.4)
2 (20.0)
7 (17.9)
1
History of adenotonsillectomy
18 (36.7)
7 (70.0)
11 (28.2)
0.025
Age mean (years)
16.2 ± 2.1
16.9 ± 2.3
16.0 ± 2.0
0.29
Height mean (cm)
169.3 – 9.9
179.0 – 5.5
166.8 – 9.2
