Obesity Surgery

, Volume 21, Issue 2, pp 200–206

The Effects of Bariatric Surgery on Asthma Severity

Authors

    • Department of Internal Medicine, Division of Pulmonary and Critical Care MedicineUniversity of Michigan Medical Center
  • Alan P. Baptist
    • Department of Internal Medicine, Division of Allergy and Clinical ImmunologyUniversity of Michigan Medical Center
  • Zhaohui Fan
    • Department of Surgery, Michigan Surgical Collaborative for Outcomes Research and Evaluation (M-SCORE)University of Michigan Medical Center
  • Arthur M. Carlin
    • Department of Surgery, Division of General Surgery, Bariatric Surgery CenterHenry Ford Health System
  • Nancy J. O. Birkmeyer
    • Department of Surgery, Michigan Surgical Collaborative for Outcomes Research and Evaluation (M-SCORE)University of Michigan Medical Center
Clinical Research

DOI: 10.1007/s11695-010-0155-6

Cite this article as:
Reddy, R.C., Baptist, A.P., Fan, Z. et al. OBES SURG (2011) 21: 200. doi:10.1007/s11695-010-0155-6

Abstract

Background

Excess body weight increases both the risk and severity of asthma. Several studies indicate that bariatric surgery decreases asthma severity, but either enrolled few patients or were not focused primarily on asthma. Furthermore, none compared the effects of different bariatric surgical procedures.

Methods

Subjects underwent bariatric surgery at member institutions of the Michigan Bariatric Surgery Collaborative between 06/06/2006 and 5/14/2009. Patient records provided data on baseline demographics, asthma medication use, comorbidities, body mass index, type of procedure and perioperative complications. One year later, patients received a follow-up mail survey covering weight and use of asthma medications at that time.

Results

Of the 13,057 bariatric surgery patients, 2,562 (18.6%) reported use of asthma medications at baseline. Several comorbidities were significantly more common in asthma patients, who also experienced significantly more perioperative wound and respiratory complications. Among 257 asthma patients who participated in a 1-year follow-up survey, 13 of 28 who had initially used oral corticosteroids for symptom control no longer required them, while use of inhaled corticosteroids decreased from 49.8% to 29.6%. Reduction in intensity of asthma therapy correlated with presence of sleep disorders and extent of weight loss on univariate analysis but not multivariate analysis. Patients who underwent laparoscopic adjustable gastric banding (LAGB), which was associated with less weight loss than other surgical modalities, were significantly less likely to reduce the intensity of their asthma therapy.

Conclusions

Bariatric surgery decreases the intensity of medication required to control patients’ asthma symptoms, although LAGB appears to produce less significant effects.

Keywords

AsthmaBariatric surgeryBody mass indexSteroidsObesityWeight loss

The prevalence of both asthma and obesity is increasing steadily. The percentage of the adult population that is obese (body mass index (BMI) ≥ 30) has more than doubled since 1960, reaching 34% in 2003–4 [1]. The increase in more extreme values of excess weight has been even greater. The prevalence of BMI ≥ 40 more than quadrupled between 1986 and 2000, affecting about 2% of all Americans [2]. By 2004, 3.6% of American women and 9.6% of African-American women were morbidly obese [3].

At the same time, the prevalence of asthma in adults has reached more than 7% [4]. More than 30 cross-sectional or case-control studies demonstrating an association of obesity with asthma prevalence suggest that these two epidemics may be related [5]. Furthermore, a number of prospective studies have shown that obesity increases asthma risk in previously asthma-free patients [5]. There is also evidence that obesity increases asthma severity [6, 7] while, conversely, severity is decreased by weight loss [8]. The evidence is particularly strong for intense dietary intervention, where three studies have shown improvements in both asthma symptoms and lung function [911] and a fourth showed that even alternate-day caloric restriction produced symptomatic improvement [12]. This improvement may reflect relief of the mechanical effects of obesity that are believed to contribute to asthma severity as well as the metabolic effects of weight loss [13].

Current guidelines indicate that bariatric surgery is appropriate for patients with BMI ≥ 40 or those with BMI ≥ 35 who have significant medical, physical, or psychological comorbidities expected to respond to weight loss [14, 15]. Most guidelines emphasize the need for an adequate trial of lifestyle modification and weight-loss drugs prior to surgery, but studies have demonstrated that long-term results with these interventions are typically disappointing [16, 17]. Only bariatric surgery appears generally capable of producing sustained weight loss.

Although previous studies have suggested that bariatric surgery may have beneficial effects on asthma, few have been rigorously performed. In particular, assessment of asthma improvement was often based entirely on patient self-report of symptomatic changes, which is notoriously subject to the placebo effect. Furthermore, none have compared the effects of different types of bariatric surgery on asthma severity. This led us to conduct a retrospective study of the effects on asthma severity of different types of bariatric surgical procedures conducted at our multi-institutional consortium, employing changes in usage of asthma medications as an objective measure of disease severity.

Methods

Study Population

This study is based on analysis of data from the Michigan Bariatric Surgery Collaborative (MBSC). As described in greater detail elsewhere [18], the MBSC is a regional consortium of hospitals and surgeons performing bariatric surgery in Michigan. The MBSC clinical registry now enrolls more than 95% of the patients undergoing bariatric surgery in the state of Michigan (approximately 5,000 patients/year). Participating hospitals submit data for all of their bariatric surgery patients.

In the MBSC, medical records are abstracted for each patient at the end of the perioperative period (30 days after surgery). The medical record reviews are performed by centrally trained nurse data abstractors using a standardized and validated instrument. Each participating hospital receives an annual on-site visit to verify the accuracy and completeness of their MBSC clinical registry data. Long-term follow-up data is obtained, via an annually mailed survey, from patients who consent to participate in that portion of the project.

For this study, we identified all patients undergoing bariatric surgery between 06/06/2006 and 5/14/2009; this represents 13,057 patients from 23 hospitals. Patients were included if they had undergone primary or revisional bariatric procedures including: open or laparoscopic Roux-en-Y gastric bypass (open or lap RYGB), laparoscopic adjustable gastric banding (LAGB), biliopancreatic diversion with duodenal switch (BPD/DS), or sleeve gastrectomy procedures.

Data Collected

We collected data on patient characteristics including demographics, preoperative medication use, medical history, and weight-related and other comorbidities. Patients were considered to have asthma if they had a clinical diagnosis of asthma recorded in the medical record during the time of their contact with the bariatric program and were currently using medication for the treatment of asthma. At baseline, asthma severity was stratified according to the type of medication required to maintain asthma control: daily oral corticosteroids, daily inhaled corticosteroids, or β-agonists on an as-needed basis. Data were also collected on 12 different types of perioperative complications, ranging in severity from wound dehiscence to death, potentially related to bariatric surgery.

Long-term outcomes include weight loss, comorbidity resolution, obesity-specific and generic quality of life, and patient satisfaction. Improvement in asthma severity was recorded if the patient indicated any decrease from baseline in the class of medications required to control asthma symptoms.

Statistical Analysis

Standard statistical methods, including chi-square tests for categorical and t tests for continuous variables, were used to compare the baseline patient characteristics and perioperative complication rates among patients with and without asthma. Logistic regression was used to evaluate multivariate predictors of asthma symptom improvement. All analyses were performed by SAS version 9.1 (Cary, NC).

Results

Patient Demographics and Comorbidities

Out of 13,057 patients, 2,562 (18.6%) reported a current diagnosis of, and medication use for, asthma. A variety of comorbidities, including gastroesophageal reflux disease (GERD), peptic ulcer disease, cholelithiasis, diabetes, cardiovascular disease, sleep apnea, musculoskeletal disorders, and psychological disorders (notably anxiety, depression, and bipolar disorder) were found to be significantly more common in patients with asthma (Table 1). Bariatric procedure type was similar between the two groups.
Table 1

Characteristics of bariatric patients with and without asthma

Characteristic

No asthma (N = 10,495)

Asthma (N = 2,562)

p value

Demographics/medical history

   

 Age >50 years (%)

37.6

38.3

0.52

 Male gender (%)

22.0

13.4

<.0001

 White race (%)a

88.3

84.4

0.08

 Private insurance (%)

74.6

68.1

<.0001

 Current smoker (%)

4.2

4.8

0.22

 BMI > 50 (%)

29.9

37.6

<.0001

 Mobility problems (%)

4.9

8.0

<.0001

 Prior VTE (%)

4.0

5.5

0.0009

Comorbidity

   

 Lung disease (%)

8.5

100.0

<.0001

 Cardiovascular disease (%)

55.1

58.6

0.001

 Hyperlipidemia (%)

49.2

52.1

0.009

 GERD (%)

45.4

58.9

<.0001

 Peptic ulcer disease (%)

2.7

3.9

0.001

 Cholelithiasis (%)

26.5

31.7

<.0001

 Urinary incontinence (%)

19.3

26.7

<.0001

 Diabetes (%)

33.0

35.7

0.01

 Liver disorder (%)

3.8

4.8

0.014

 Sleep apnea (%)

42.7

51.3

<.0001

 Musculoskeletal disorder (%)

73.6

80.1

<.0001

 History of hernia repair (%)

3.1

4.0

0.02

 Psychological (%)

45.0

56.1

<.0001

Bariatric procedure

   

 Laparoscopic RYGB (%)

52.7

53.1

0.71

 LAGB (%)

35.9

34.6

0.21

 Open RYGB (%)

5.2

6.5

0.009

 BPD/DS (%)

2.0

2.5

0.11

 Sleeve gastrectomy (%)

4.2

3.4

0.06

aThis variable only available for patients that consent to participate in the follow-up study

Of the 2,562 patients with asthma, 1,446 consented to contact 1 year following surgery. Six hundred six consenting patients with asthma had reached the 1-year follow-up point at the time of analysis, of whom 257 (42.4%) returned the survey. The demographic information for these patients is listed in Table 2 along with data on BMI and excess body weight at baseline and at 1-year follow-up. Weight data demonstrate the effectiveness of bariatric surgery, with mean loss of excess body weight during the year following bariatric surgery being 56%.
Table 2

Demographics and weight loss among asthma patients at 1-year follow-up (N = 257)

Gender (N, %)

 

 Female

228 (88.7)

 Male

29 (11.3)

Race (N, %)

 

 White

217 (84.4)

 Black

27 (10.5)

 Other

13 (5.1)

Income (N, %)

 

 ≤$45,000/year

116 (45.1)

 >$45,000/year

132 (51.4)

 Unknown/refused

9 (3.5)

Mean age (year)

48 ± 10.3

Mean baseline BMI

49 ± 10.0

Mean BMI at follow-up

35 ± 8.1

Mean baseline excess body weight (lb)

156 ± 57.7

Mean excess body weight at follow-up (lb)

70 ± 46.0

Mean percentage of excess body weight loss (%)

56 ± 22.9

Procedures and Complications

Somewhat over half (58.8%) of the patients available for 1-year follow-up underwent lap RYGB and another 27.6% were treated using LAGB. Surgical techniques for the remainder were split between open RYGB and BPD/DS. Asthma patients were significantly more likely to experience wound complications and, as might be expected, respiratory complications following the procedure, resulting in a greater frequency of Grade 1 complications (Table 3). No significant differences were observed between patients with and without asthma in the frequencies of other complications or of more severe complications. Compared to patients without asthma, asthma patients had a somewhat longer hospital stay (2.1 vs. 1.9 days; p = 0.0004) and during the first 30 days following surgery were more likely to require emergency department treatment (10.3% vs. 6.8%; p < 0.0001) or rehospitalization (5.4% vs. 4.2%; p = 0.006).
Table 3

Peri-operative complications among bariatric patients with and without asthma

Characteristic

No asthma (N = 10,495)

Asthma (N = 2,562)

p value

Complications

   

 Bowel obstruction (%)

0.52

0.63

0.49

 Abdominal abscess (%)

0.62

0.78

0.36

 Wound complication (%)

2.87

4.37

<.0001

 Hemorrhage (%)

1.46

1.91

0.09

 VTE (%)

0.44

0.51

0.64

 Cardiac (%)

0.10

0.16

0.40

 Renal failure (%)

0.30

0.47

0.20

 Respiratory (%)

1.08

2.00

0.0002

 Anastomotic problem (%)a

1.46

2.04

0.09

 Leak (%)*

0.95

0.66

0.25

 Band slippage (%)b

0.11

0.11

0.96

 Port site infection (%)b

0.53

0.34

0.46

 Band outlet obstruction (%)b

0.24

0.00

0.15

 Gastric perforation (%)b

0.08

0.11

0.76

Complication severity

   

 No complication (%)

92.57

89.85

<.0001

 Grade I (%)

4.62

6.83

<.0001

 Grade II (%)

2.46

2.81

0.31

 Grade III (%)

0.35

0.51

0.26

Adverse events

   

 Reoperations (%)

1.94

1.89

0.84

 Extended LOS (%)

5.33

7.57

<.0001

 Readmissions (%)

4.18

5.43

0.006

 Transfers (%)

0.17

0.31

0.15

 ER visits (%)

6.75

10.30

<.0001

 Deaths (%)

0.14

0.12

0.75

 Any adverse event (%)

13.10

17.68

<.0001

aThis variable is only available for patients who had lap RYGB, open RYGB, BPD/DS or Sleeve gastrectomy

bThese variables are only available for patients who had LAGB

Weight Loss and Reductions in Asthma Severity

As shown in Table 4 and Fig. 1, the number of patients regularly using oral corticosteroids, daily inhaled corticosteroids, or inhaled bronchodilators to control their asthma all decreased significantly at 1-year follow-up. One hundred-one of these patients became medication-free. There were no significant changes in the relatively small numbers of patients using anti-leukotrienes.
Table 4

Asthma severity at baseline and follow-up (N = 257)

Asthma severity

Baseline

Follow-up

p value

No medication

0 (0.0%)

101 (39.3%)

<0.0001

Inhaled bronchodilators

90 (35.0%)

52 (20.2%)

<0.0001

Anti-leukotriene (Singulair, Accolate, or Zyflo)

11 (4.3%)

13 (5.1%)

0.835

Inhaled corticosteroids

128 (49.8%)

76 (29.6%)

<0.0001

Oral steroids

28 (10.9%)

15 (5.8%)

0.055

https://static-content.springer.com/image/art%3A10.1007%2Fs11695-010-0155-6/MediaObjects/11695_2010_155_Fig1_HTML.gif
Fig. 1

Changes in asthma medication usage among patients available for 1-year follow-up. Figure shows the percentage of patients in the follow-up group who were receiving each type of maintenance asthma medication at baseline and at the time of follow-up

Univariate analysis of factors associated with improvement in asthma severity following surgery (Table 5) showed an association between the type of surgical procedure as well as amount of weight loss and ability to decrease the class of medication required to control asthma symptoms (p = 0.04). Less predictably, those with sleep disorders at baseline were likewise more likely to decrease medication use (p = 0.04). No significant effects of age, race, gender, income or baseline BMI were observed, nor was asthma outcome affected by the presence of GERD, psychological disorders, or current or past smoking history. On multivariate analysis, type of surgery remained significant (p = 0.007), with those who underwent LAGB being only 37% as likely to demonstrate improvement in asthma severity. Patients who underwent LAGB surgery also lost a smaller percentage of their excess body weight (mean, 38%) than those who underwent other procedures (means, 58–76%). The presence of sleep disorders (p = 0.08) and extent of weight loss (p = 0.91) no longer remained significant on multivariate analysis.
Table 5

Factors associated with improvement in asthma severity after bariatric surgery

Factor

OR

95% C.I.

p value

LL

UL

Univariate analysis

    

Age <50

0.91

0.55

1.52

0.73

Male

1.29

0.57

2.93

0.54

White/Black/Other

    

 White vs. Other

3.04

0.89

10.41

0.08

 Black vs. Other

2.91

0.70

12.1

0.14

Current smoker

2.85

0.31

25.8

0.36

Former smoker

0.99

0.59

1.67

0.98

Income ≤$45,000

1.25

0.75

2.07

0.40

Baseline BMI

1.01

0.99

1.04

0.42

Decrease in BMI

1.04

1.00

1.08

0.04

Psychological disorder

0.67

0.40

1.13

0.13

GERD

0.67

0.40

1.13

0.13

Sleep disorder

1.74

1.04

2.91

0.04

Type of surgery (LAGB vs. all others)

0.37

0.21

0.65

0.0006

Multivariate analysis

    

Decrease in BMI

1.00

0.95

1.05

0.91

Sleep disorder

1.60

0.94

2.72

0.08

Type of surgery (LAGB vs. all others)

0.37

0.18

0.76

0.007

Discussion

We find that weight loss due to bariatric surgery profoundly improves asthma symptoms. At 1 year following surgery, many patients have been able to eliminate the class of medication initially required for adequate control in favor of one less aggressive; nearly 40% are off maintenance medications entirely. A key finding is that this reduction in medication intensity is less likely for patients who underwent LAGB than for those treated with other surgical modalities. It is possible that the lesser effect of LAGB surgery on asthma severity reflects the smaller weight reduction seen with this technique. Extent of decrease in BMI significantly predicted reduction in asthma medication use on univariate analysis, but this effect no longer appeared on multivariate analysis. The implication is that weight loss is confounded by one or both of the two remaining significant or nearly significant predictors, type of surgery and sleep disorders. Since the least effective type of surgery is also associated with the least weight loss, this suggestion appears plausible. We note, however, that the statistical significance of weight loss may be regained as more patients reach the 1-year mark and are added to the database for analysis. Other significant factors may also be found as the database grows.

Why presence of sleep disorders might be associated with greater improvement in asthma control following bariatric surgery is not clear. It may be possible to speculate that the presence of sleep disorders signals obesity-associated mechanical effects on ventilation, relief of which following weight loss contributes to respiratory improvement. It has also been speculated that sleep apnea and consequent sleep deprivation may lead to overeating and weight gain, inhibition of which could render bariatric surgery more effective, although this has not been confirmed in controlled studies [19].

Our data do not directly address the impact on medical resource utilization of bariatric surgery among patients with asthma, a matter of interest in the context of growing concern about overall healthcare costs. Medical resource utilization has been shown to correlate with disease severity [2022], however, and medication usage is a validated surrogate marker for severity [23]. Our demonstration of a reduction in intensity of drug therapy thus implies a reduction in medical resource utilization, although the extent of the reduction cannot be specified.

A number of previous studies have demonstrated favorable effects of bariatric surgery on asthma severity. Our study is, however, the first to explicitly compare the effects of different bariatric surgical procedures and thus the first to show that a specific modality, LAGB, has less effect on asthma outcomes than other techniques. Previous studies of asthma in the context of bariatric surgery have either been restricted to patients undergoing a specific type of surgery [2432] or have analyzed all types of bariatric surgery as a group [33, 34].

In many of the previous studies [27, 28, 30, 31], asthma severity was assessed by patient self-report, sometimes supplemented by clinical assessment. As is well known, both patient and physician assessment are highly subject to the placebo effect. One study used insurance reports of physician office visits in which asthma was recorded as the primary presenting diagnosis [33]. As the authors note, the recorded primary diagnosis may be somewhat arbitrary for patients with multiple morbidities. Four previous studies have assessed severity by changes in asthma medication usage [25, 29, 32, 34], as we do. The most thorough previous assessment of the effect of bariatric surgery on asthma severity was that of Maniscalco and colleagues [26]. This was a small prospective study that found improvements in both asthma symptoms and objective lung function in bariatric surgery patients but not in non-operated controls. These results complement ours by showing that the improvement in asthma and pulmonary function is due to bariatric surgery rather than to routine management of asthma and obesity without associated weight loss.

The high proportion of female patients in our cohort (79.6%) is likely to reflect the substantially greater prevalence of obesity in women. It is also possible that obese females are more likely than males to choose bariatric surgery. However, asthma statistics also show that, among adults, women are almost twice as likely to have asthma as are men [4]. These similar gender patterns for obesity and adult asthma suggest that common mediators may play a role in both conditions.

Our results also show that asthmatic patients are more likely than those without asthma to suffer respiratory and wound complications in the perioperative period. While the respiratory complications might be somewhat expected, the wound complications are not. We also find that, at baseline, patients with asthma had higher rates of a variety of comorbidities, including GERD, sleep apnea, diabetes, and others. Further research is warranted on possible common mechanisms underlying asthma, other chronic diseases, and wound complications.

This was a retrospective study, which is best suited to showing correlation rather than causation. While univariate analysis showed that improvement in asthma severity correlated with weight loss, this favorable effect may be indirect. It is likely that weight loss also improves comorbidities such as sleep disorders and GERD, and the improvement in these comorbidities may well account for at least some of the improvement in asthma severity.

Our results may also be affected by loss-to-follow-up bias. About 40% of patients eligible for 1-year follow-up returned the survey form, which is typical for such surveys. Response rate could nevertheless have been affected by overall satisfaction with bariatric surgery, although differences specifically in asthma severity would seem unlikely to have any effect. It is also conceivable that patients who withheld consent for 1-year follow-up differed in relevant ways from those who consented, although there is no specific reason to believe this might have been the case.

In summary, our results show that asthma severity is influenced by body weight and that bariatric surgery can reduce asthma severity. The mechanisms involved remain unclear, however, and may not solely reflect extent of weight loss. LAGB appears to be less effective for this purpose than other surgical techniques. In total, our results indicate that asthma is a comorbidity potentially justifying bariatric surgery in patients with a BMI ≥ 35.

Conflict of Interest

The authors declare that they have no conflict of interest.

Copyright information

© Springer Science + Business Media, LLC 2010