Discontinuation and reduction of asthma medications after metabolic and bariatric surgery: A systematic review and meta-analysis

Summary

Obesity is a risk factor for asthma. Metabolic and bariatric surgery (MBS) is a safe and effective treatment option for obesity. Weight reduction via MBS, in turn, may improve asthma outcomes and decrease the need for asthma medications. The aim of the systematic review and meta-analysis is to explore the available evidence focused on the impact of MBS on the improvement of asthma outcomes via the discontinuation and reduction of asthma medications. After a comprehensive search in the PubMed, Embase, and Cochrane Central Register of Controlled Trials (CENTRAL) databases, 15 studies, including pre–post MBS data on asthma medication use among adults, were eligible for the systematic review. Thirteen studies reported the proportion of patient who discontinued asthma medication post-MBS and was meta-analyzed using random effects. Results showed 54% patients completely discontinued asthma medications (95% confidence interval 42%–67%, I² = 86.2%, p < 0.001). The average number of asthma medications was also decreased by approximately 22%–46%. MBS provides strong therapeutic benefits for patients with asthma, as evidenced by the complete discontinuation of asthma medications in over 50% of MBS completers. The inference was limited by the small number, variations in follow-up time and rates, and heterogeneity of studies. Studies that include more ethnically diverse participant samples are needed to improve generalizability.

1 |. INTRODUCTION

According to the World Health Organization (WHO), obesity is defined as excessive fat accumulation that may negatively impact an individual’s health. Body mass index (BMI) is a commonly used anthropometric measure to characterize obesity and is calculated by dividing one’s weight (kilograms) by their height (meters) in meters squared (kg/m²).¹ Individuals with a BMI of 30 kg/m² or greater are considered to have obesity.² Approximately 13% of the world’s population 18 years and older have obesity.¹ Obesity is a major public health problem in high-resource countries in particular. For instance, 42.4% of the US adult population were characterized as having obesity between 2017 and 2018.³

Asthma, a chronic respiratory disease, is a prevalent comorbidity related to obesity affecting 4.3% of patients worldwide and 8% of adults in the United States.^4,5 The exact cause of asthma is unknown, and it cannot be cured in most cases.⁵ Therefore, medications are commonly prescribed to treat asthma. Without appropriate management, asthma can lead to frequent emergency department visits, hospitalizations, and even mortality.⁵

Increased BMI is a risk factor for asthma. It was suggested that adipose tissue releases proinflammatory factors, including interleukin (IL)-6, that in turn, worsen asthma control. In addition, immune system-related allergic responses may be suppressed in patients with obesity and lead to increased asthma airway inflammation.⁶ Specifically, adults with Class I obesity (BMI between 30 and <35 kg/m²) have a 1.4 times higher risk of asthma compared with adults with normal weight. Additionally, adults with Class III obesity (BMI of 40 kg/m² or higher), specifically those with a BMI greater than 50 kg/m² have a 2.5 times higher risk of asthma compared with adults with normal weight.⁶ However, in some cases, asthma may occur before the onset of obesity.⁶ Regardless of whether asthma precedes obesity or whether obesity precedes asthma, weight reduction can help asthma control or may even lead to resolution.⁷ Metabolic and bariatric surgery (MBS) is one of the most effective treatment options for obesity. MBS is more effective than conventional methods of weight loss such as a reduced calorie diet and increased physical activity.⁸ Patients who complete MBS will lose on average 31% of total body weight in the first year after surgery and 24% of their total body weight within 5 years after surgery, which is significantly more compared with nonsurgical patients.⁹

To date, changes in medication use related to obesity-related comorbidities including hypertension and Type 2 diabetes following MBS have been studied^10,11; these studies suggested a substantial reduction in medication use in a large proportion of patients.^10,11 One systematic review reported pulmonary function of asthma patients improved post-MBS.¹² However, to our knowledge, no recent reviews have meta-analyzed the effect of MBS on the use of asthma medications. Understanding the extent of discontinuation and reduction of asthma medications following MBS will not only inform patients with asthma of the potential benefits of MBS but also quantify the improvement of asthma through weight loss. Thus, the aim of the systematic review and meta-analysis is to explore the available evidence on the impact of MBS on asthma outcomes via (1) the complete discontinuation of asthma medication and (2) the reduction of the number of asthma medications.

2 |. METHODS

2.1 |. Study protocol registration

This study’s protocol has been registered at PROSPERO (CRD42022308496), a database of prospectively registered systematic reviews funded by the National Institute for Health Research (NIHR) of the United Kingdom (https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42022308496). This report follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guideline.

2.2 |. Search strategy and eligibility criteria

Databases including PubMed, Embase, and Cochrane Central Register of Controlled Trials were used to search a combination of search terms including “asthma medication,” “antiasthmatic,” “anti-asthmatic,” or “bariatric surgery,” “metabolic and bariatric surgery,” “weight loss surgery,” “obesity surgery,” “gastric bypass,” “sleeve gastrectomy,” “gastric band,” or “biliopancreatic diversion with duodenal switch.” The search terms were modified for different databases. Studies published between 01/01/1990 and 02/01/2022 were included for initial screening. Duplicates were manually identified and excluded. The main inclusion criterion was original articles focusing on asthma medication usage before and after MBS. Studies were excluded if they were protocols, case reports, nonhuman studies, reviews, or stand-alone abstracts.

2.3 |. Study selection and data extraction

Title and abstracts of identified records were screened independently by two reviewers (LX and AC) before the further full-text screening. Differences between the two reviewers were assessed, reviewed, and resolved by a senior team member (SEM). All final full articles were successfully obtained. Relevant data on authors, country, study type, participants, intervention, and outcomes were extracted into a standardized data extraction form.

2.4 |. Data synthesis and analysis

Data were analyzed based on the two types of prior specified outcomes: (1) discontinuation of asthma medication and (2) change of asthma medication before and after bariatric surgery, respectively. Specifically, the proportion of those discontinuation asthma medications post-surgery was meta-analyzed using the metan command in STATA Version 15.1, specifying the inverse variance approach for pooling the logit of the proportion. This statistical package accommodates extreme values (such as an observed proportion of “1” in one study). The heterogeneity of the effect sizes was tested by using the I² test for homogeneity. Since I² = 86.2%, p < 0.001, a DerSimonian and Laird random-effects meta-analysis was reported.¹³ The final findings were displayed in a forest plot of the proportions with the pooled effect size with 95% confidence interval (95% CI).

There were only four studies that reported the change in the number of asthma medications following MBS; hence, a meta-analysis was not performed due to lack of statistical power. Instead, the summary of changes in asthma medications was synthesized in a table stratified by different time points post-MBS.

2.5 |. Risk of bias assessment

The National Institutes of Health (NIH) standardized study quality assessment tools were used to evaluate the quality of the present study.¹⁴ Publication bias for discontinuation of asthma medications was assessed with a visual inspection of funnel plots.

2.6 |. Sensitivity analysis

Leave-one-out analysis was conducted to assess the influence of a single study on the overall effect of the meta-analysis by omitting one study each time, which can help identify the source of heterogeneity. Our sensitivity analysis suggested all included studies are within the 95% CI of the original pooled random effects except for one study (Reddy et al, 2011),¹⁵ which slightly increased the proportion after excluding (Figure S1).

3 |. RESULTS

The study selection process applying the PRISMA guideline is demonstrated in Figure 1. The search first yielded a total of 380 records (318 records in PubMed, 15 in Cochrane, and 47 from Embase). Seventeen duplicate records and 11 conference abstracts were removed, so 352 records have remained for the title and abstract screening. After the exclusion of 322 abstracts that did not mention asthma or MBS, a total of 30 full-text articles were retrieved and reviewed for eligibility. Thirteen of these were excluded due to no medication utilization information. In addition, one review and one study protocol were also excluded. Finally, there were 15 publications that met the prespecified inclusion criteria and were included in the systematic review.

FIGURE 1.

PRISMA flow diagram

3.1 |. Study characteristics

Most studies were conducted in the United States (n = 9),^15–23 followed by other high-resource countries such as Australia (n = 3),^24–26 Canada (n = 1),²⁷ Norway (n = 1),²⁸ and New Zealand (n = 1).²⁹ The majority were cohort studies including seven prospective cohort studies^19–25 and six retrospective cohort studies.^15–19,28 There were one cross-sectional study²⁶ and one controlled clinical trial.²⁷ The study sample of each study ranged from 12 to 13,507 with 60% of the studies having more than 100 participants. In regard to MBS procedure type, 40% of the studies investigated the outcomes by various procedures, 33.3% were only conducted in patients who had laparoscopic adjustable gastric banding (LAGB), 20% did not specify, and 20% included gastric bypass (Table 1).

TABLE 1.

Population characteristics of included studies

Study	Country	Data type	Surgery type	Total N	Age (SD)	Female
Controlled clinical trial
Boulet et al, 2012	Canada	Prospective data collection	Not specified	12	41 (10)	75%
Cohort study
Guerron et al, 2019a	USA	Single institution's database	LRYGB (79%); LSG (10.7%); LAGB (8.1%); duodenal switch (2.3%)	751	46.8 (11.6)	87.70%
Abbas et al, 2015a	USA	Medical records of a single hospital	LSG (n = 30); LRYGB (n = 53)	98	63.4 (3.1)	80%
Hewitt et al, 2014a	Norway	Medical records of a single hospital	Gastric bypass (n = 101), biliopancreatic diversion with duodenal switch (n = 10), LSG (n = 2)	140	40 (9)	71%
Sikka et al, 2010a	USA	Michigan health maintenance organization	Not specified	320	45.6 (9.3)	85.60%
Reddy et al, 2011a	USA	Michigan Bariatric Surgery Collaborative	RYGB, LAGB, BPD/DS, and sleeve gastrectomy	13,057	48 (10.3)	86.60%
Omana et al, 2009a	USA	Medical records of a single hospital	LSG (n = 49); LAGB (n = 74)	123	LSG: 45 (12); LAGB: 41 (14)	76.40%
Ahroni et al, 2005	USA	Prospective data collection at a single institution	LAGB	195	43.8 (10.1)	82.80%
Spivak et al, 2005	USA	A private US hospital setting	LAGB	500	42 (range 18–63)	86.40%
O'Brien et al, 2002	Australia	Prospective data collection at a single institution	LAGB	709	41 (range 16–71)	85.00%
Dhabuwala et al, 2000	New Zealand	Prospective data collection at a single institution	Silastic ring gastric bypass	157	42 (range 15–62)	77.07%
Dixon et al, 1999	Australia	Prospective data collection at a single institution	LAGB	33	39 (range 17–69)	88%
Murr et al, 1995	USA	Prospective data collection at a single institution	Not specified	62	57 (1)	79%
Macgregor et al, 1992	USA	Medical records of a single hospital	Gastric bypass (n = 35); vertical banded gastroplasty (n = 5)	40	45 (range 23–68)	80%
Cross-sectional
Keating et al, 2013	Australia	Australia Medicare	LAGB	9542 in primary analysis; 11,769 in validation analysis	60.7% 35–54 years	77.70%
Controlled clinical trial
Boulet et al, 2012	Not specified	51.2 (7.3) kg/m2	34.4 (4.3) kg/m2	The effects of bariatric surgery on asthma in severely obese subjects	1 year	83.3%
Cohort study
Guerron et al, 2019a	Caucasian 47 (58.8%); African American 29 (36.2%); Other 4 (5.0%)	49 (8.2) kg/m2	34.5 (7.5) kg/m2	Utilization of asthma medications at baseline, 30 days, 60 days, 6 months, 1 year, 2 years, and 3 years after MBS	3 years	<69%e
Abbas et al, 2015a	Not specified	47 (7.9) kg/m2	Not reportedb	The safety and efficacy of bariatric surgery in patients > 60 years old	4 years	84.7%
Hewitt et al, 2014a	Not specified	47.4 (6.3) kg/m2	Not reportedc	Pulmonary function and asthma and obstructive sleep apnea syndrome before and 5 years after bariatric surgery	5 years	80.7%
Sikka et al, 2010a	African American (62%)	49.0 (6.4) kg/m2	34.1 (6.1) kg/m2	Respiratory symptoms after bariatric surgery	2 years	100%
Reddy et al, 2011a	White (84.4%)	49 (10.0) kg/m2	35 (8.1) kg/m2	Asthma symptom improvement after bariatric surgery at 1-year follow-up	1 year	42.4%
Omana et al, 2009a	Not specified	LSG: 52 (11) kg/m2; LAGB: 44 (5) kg/m2	LSG: 37.8 (5.5) kg/m2; LAGB: 36 (1.7) kg/m2	Compare the rates for resolution of common comorbidities between LSG and LAGB	3 years	100%
Ahroni et al, 2005	Not specified	45.8 (7.7) kg/m2	32.3 (7) kg/m2	Weight loss, change in comorbidities, medication usage, and general health status after bariatric surgery	1 year	77.2%
Spivak et al, 2005	Not specified	45.2 (12.8) kg/m2	34.9 (N/A) kg/m2	Weight loss and improvement of obesity-related illness after bariatric surgery	3 years	32.6%
O'Brien et al, 2002	Not specified	45.0 (7) kg/m2	31 (N/A) kg/m2	Medium-term effects of bariatric surgery on weight, health and quality of life	1 year	98.6%
Dhabuwala et al, 2000	Not specified	45 (range 33–97) kg/m2	28 (range 20–52) kg/m2	Changes of co- morbidconditions following bariatric surgery	5 years	90%
Dixon et al, 1999	Not specified	45.1 (7) kg/m2	32.9 (4.7) kg/m2	The changes in asthma after LAGB	4 years	96.9%
Murr et al, 1995	Not specified	Not reported	Not reportedd	The safety and efficacy of bariatric surgery in patients >50 years old	5 years	100%
Macgregor et al, 1992	Not specified	46 kg/m2	30 kg/m2	Effect of bariatric surgery on asthma in morbidly obese patients	4 years (range 211 years)	100%
Cross-sectional
Keating et al, 2013	Not specified	Not reported	Not reported	Pharmaceutical utilization and costs before and after bariatric surgery	4 years	100%

Abbreviations: BPD/DS, biliopancreatic diversion with duodenal switch; LAGB, laparoscopic adjustable gastric banding; LRYGB, laparoscopic Roux-en-Y gastric bypass; LSG, laparoscopic sleeve gastrectomy; RYGB, Roux-en-Y gastric bypass.

^aRetrospective cohort study.

^bThe percent total body weight loss was 26.9% (SD = 9.6) at the 1-year follow-up.

^cWeight loss was 42 kg (31% of total body weight) in women and 38 kg (24%) in men.

^dPercent excess body weight loss was 33% (SD = 6).

^eAbout 69% of patients had complete data up to 1-year post-operation; the follow-up rate at 3 years was not reported.

All studies were conducted in the adult population and included predominately females. The average age of the participants ranged from 39 to 63.4 years, and 71%–88% were females. Most studies (80%) did not specify ethnicity. The three studies that mentioned ethnicity were conducted in the United States, and one study reported 84.4% were non-Hispanic White, whereas the remaining two US studies had a more ethnically diverse sample. Presurgical BMI was not specified in two studies (one used Australia Medicare data,²⁶ and another one was conducted in 1995²²); other studies reported the average presurgery BMI ranged from 45 to 52 kg/m². The average post-surgery BMI decreased from 28 to 37.8 kg/m², and other studies reported the percent total body weight loss ranged from 24% to 33% (Table 1).

The majority of included studies primarily aimed to examine the effects of MBS on asthma outcomes, and only three studies specified the primary aim was to assess medication utilization before and after surgery. The study length varied from short term (e.g., 1 year) to long term (e.g., 5 years) with the most reporting a follow-up time of 3 or 4 years. The follow-up rate was either reported by the authors^{15,16,18–20,22–24,26,29} or calculated by using the number of patients at the end of follow-up divided by the number of patients using asthma medications at the baseline.^{17,21,25,27,28} Most studies (N = 11) had follow-up rates greater than 80%; however, Reddy et al¹⁵ conducted a mail-based survey after 1 year of bariatric surgery and had a response rate that was less than 50% (257 patients out of 606 consented patients finished mail-based surveys). The low response rate may explain the influential results of the leave-one-out analysis due to the missingness of asthma medication information post-MBS (Table 1).

3.2 |. Discontinuation of asthma medications

Thirteen out of 15 articles (total N = 982) reported the proportion of discontinuation of asthma medications after MBS and had available data for meta-analysis.^{15,17–25,27–29} Figure 2 shows a forest plot of the proportion of patients who discontinued taking asthma medications after MBS. Although there was significant heterogeneity in the reported proportions of patients that discontinued asthma medications (I² = 86.2%, p < 0.001), a random-effects meta-analysis indicated the proportion of patients that discontinued asthma medications is 54% (95% CI 42%–67%). And this proportion is significantly different from zero. The funnel plot illustrates the publication bias of included studies is likely to be minimal because effect sizes appear symmetrical around the vertical line denoting the pooled effect size (Figure S2).

FIGURE 2.

Proportion of asthma medications discontinuation after bariatric surgery

3.3 |. Change of asthma medications

Table 2 illustrates the changes in the mean count of asthma medications before and after MBS. Due to the variability of reporting the outcome, heterogeneity in measures, and the limited number of studies, it was not possible to pool the data for meta-analysis. However, a qualitative analysis was conducted to summarize the changes in terms of the number of asthma medications. Guerron et al¹⁶ reported the mean count of medications and rate ratios at presurgery, 30 days, 6 months, 1 year, and 3 years post-surgery, respectively. The rate ratio decreased over time and was 0.54 (95% CI 0.45–0.65) at the 3-year follow-up. Hewitt et al²⁸ reported the mean count of asthma medications decreased from 2.2 to 1.4 after 5 years post-MBS. Similarly, Sikka et al¹⁸ found the mean count dropped about half from 7.0 to 3.8 after 1-year post-MBS. Keating et al²⁶ used Australia Medicare data and reported the incidence rate per person/year of using asthma medications was 0.59 before surgery but declined to 0.46 after 2 years post-MBS. The calculated rate ratio was 0.78 (0.75–0.81).

TABLE 2.

Changes in the mean count of asthma medications before and after bariatric and metabolic surgery

Study	Measures	Pre-surgery	30 days post-surgery	6 months post-surgery	1 year post-surgery	2 years post-surgery	3 years post-surgery	5 years post-surgery
Guerron et al, 2019	Mean count (SD)	1.4 (0.6)	1 (0.8)	0.9 (0.8)	0.8 (0.8)	N/A	0.8 (0.9)	N/A
	Rate ratio (95% Cl)	1.0	0.73 (0.66–0.80)	0.63 (0.56–0.70)	0.56 (0.50–0.63)	N/A	0.54 (0.45–0.65)	N/A
Hewitt et al, 2014	Mean count (SD)	2.2 (N/A)	N/A	N/A	N/A	N/A	N/A	1.4 (N/A)
Sikka et al, 2010	Mean count (SD)	7.0 (6.9)	N/A	N/A	3.8 (6.1)	N/A	N/A	N/A
Keating et al, 2013	Incidence rate (per person/year)	0.59	N/A	N/A	0.45	0.46	N/A	N/A
	Rate ratios (95% Cl)	1	N/A	N/A	N/A	0.78 (0.75–0.81)	N/A	N/A

4 |. DISCUSSION

This is the first study to meta-analyze the change in asthma medication utilization post-MBS. According to a comprehensive search strategy and mindful data extraction, 15 studies were included in the systematic review, and 13 were eligible for the meta-analysis. Studies presented in this systematic review illustrate how MBS improves asthma outcomes via the discontinuation and reduction of asthma medications. The main finding of the meta-analysis is that more than half (54% [95% CI 42%–67%], I² = 86.2%, p < 0.001) of patients were able to stop all asthma medications following MBS and achieved complete remission. Additionally, the number of asthma medications decreased from 22% to 46% based on studies that reported rate ratios.

Our results align with previous systematic reviews, but from different perspectives, that asthma outcomes significantly improved after MBS in terms of pulmonary function and quality of life.¹² For instance, two meta-analyses included six and seven studies, respectively, found a substantial increase in forced vital capacity and forced expiratory volume-one second percent in asthmatic patients after MBS.¹² Quality of life also improved in asthma patients with obesity.¹² The decrease in the number of asthma medications appears to be associated with improved pulmonary function^27,28; however, no included studies have directly measured the impact of the change in asthma medication utilization on quality of life, highlighting the need for further research in this area.

Most of the eligible studies were prospective cohort studies.^{15–25,28,29} There was one controlled clinical trial aiming to examine the effects of MBS on asthma in subjects with severe obesity.²⁷ In this non-randomized non-blinded controlled trial, authors enrolled 12 patients with asthma and 11 patients without asthma as controls prospectively. Although medication usage was comprehensively reported, the primary outcome for this study was airway responsiveness to methacholine at 12 months after surgery. In addition, one population-based cross-sectional study not only looked at medication utilization but also assessed cost change after MBS in Australia.²⁶ The decrease in the utilization rate of asthma medications was associated with a drop in annual costs from $33.72 to $23.69 per person.²⁶

A research gap on race/ethnicity was identified in this review. It is concerning that only three out of 15 eligible studies reported the inclusion of ethnic minorities. However, the prevalence of obesity is greatest among non-Hispanic Blacks (49.6%) than in other ethnic groups (44.8% in Hispanics, 42.2% in non-Hispanic Whites, and 17.4% in non-Hispanic Asians, respectively) in the United States.³ And the difference is even more concerning for women with severe obesity, which comprised a large proportion of patients who completed MBS. Specifically, 13.8% non-Hispanic Black females had severe obesity compared with 9.3% of non-Hispanic White females.³⁰ Most importantly, ethnical minorities, such as non-Hispanic Blacks, disproportionately suffer from asthma. In 2017–2019, 10.6% of non-Hispanic Blacks had asthma compared with 7.7% of non-Hispanic Whites.⁵ Current evidence has shown that ethnical minorities are less likely to achieve asthma remission than non-Hispanic White patients.³¹ Additionally, females have lower rates of asthma remission compared with males.³² Hence, inclusion and reporting of ethnically diverse males and females are highly recommended for future studies.

Furthermore, all studies were conducted in high-resource Western countries. Given the mortality of asthma is much higher in lower-income countries due to air pollution, environmental triggers, and limited access to healthcare,^33,34 and an increasing number of people have severe obesity globally,¹ more studies from developing countries are needed to help comprehensively understand the efficacy and safety of MBS on asthma medication use worldwide.

The pathophysiology behind obesity and asthma lung function has received some attention. Obesity contributes to changes in the mechanical properties of both the lungs and chest wall as a result of increased adiposity in the abdominal cavities and the mediastinum.³⁵ These changes, thus, diminish the function of the lungs, chest wall, and respiratory system overall. Specifically, the diminished function of the respiratory system changes one’s breathing pattern.³⁵ It is known that the influx of air into the lungs takes place on a negative pressure gradient. Among individuals with obesity, the change in breathing pattern occurs because both intra-abdominal and pleural pressures increase due to an increase in fat in the abdominal and thoracic cavities, which leads to a significant decrease in the expiratory reserve volume and the resting volume of the lung.³⁵ Therefore, significant weight loss following bariatric surgery will help improve asthma outcomes. However, our meta-analysis results showed that 46% of patients did not achieve asthma remission; but the weight loss outcome was very similar across studies. So, future research could focus on investigating those “hard-to-cure” asthma patients after weight loss, which may help understand other potential factors that contribute to this phenomenon.

4.1 |. Limitations and strengths

There are some limitations that should be considered when interpreting our results. First, heterogeneity existed for the pooled randomeffects estimates from the meta-analysis. The variations in follow-up time, study population, and study design may partially explain the observed heterogeneity. Second, asthma medication usage was not the primary outcome for some included studies, which may lead to reporting bias. However, publication bias as shown in the funnel plot was found to be minimal. Third, the generalizability of the study results may be limited by the small sample size and variations in the follow-up time and rates of included studies. Fourth, unmeasured confounding factors from included studies cannot be overlooked. Despite those limitations, this is the first systematic review and meta-analysis in the literature that quantified the change in asthma medication use after MBS, and the risk of publication bias was low.

4.2 |. Future directions

Using a comprehensive search in PubMed, Embase, and Cochrane Central Register of Controlled Trials, the number of studies that reported asthma medication outcomes after MBS is scarce when compared with other studies focused on other comorbidities. For example, there were at least 60 studies reported antihypertensive medication utilization before and after MBS.¹⁰ Furthermore, only two eligible studies^15,26 were population based, whereas the majority of studies with pre- and post-MBS asthma medication information had a relatively small sample size. This reveals there is an opportunity for more large-scale population-based studies focused on the impact of MBS on the utilization of asthma medication in the future.

Specific asthma medication regimens were seldomly reported. Sikka et al¹⁸ found that after MBS, the filled prescription number of β agonists, inhaled, and oral corticosteroids were all decreased by about half compared with before MBS (all p < 0.05). However, because most studies did not specify asthma medication type, but rather reported it as any asthma medication use, it is possible that the differences in asthma medications in each study may confound our study results. Hence, future studies should report detailed information on the asthma medication regimen and examine the effect of MBS on rescue and maintenance therapy of asthma, respectively.

Although the resolution and improvement of asthma may be associated with direct weight loss from MBS,^{15,17,22,24,27–29} it is difficult to discern whether the reduction in asthma medication use is a main effect of weight loss, or whether this reduction in medication use happens due to other causes. ^16,25 For example, one must consider the improvement in post-MBS dietary intake and physical activity may mediate improvements in asthma outcomes.^18,23 It is also likely that the reduction of reflux and correction of a hiatus hernia during MBS may also improve asthma-related outcomes but needs further investigation.

In addition, as suggested by previous studies, future investigations can examine if the impact of MBS on asthma medications differs by asthma severity, gender, and race/ethnicity, all potentially important effect modifiers.^15,31,32

4.3 |. Conclusion

MBS improved asthma outcomes as evidenced by a significant decrease in medication utilization, which not only leads to asthma remission in some patients but also results in a decrease in healthcare utilization and cost. In the future, large, prospective studies that include ethnically diverse participants are needed to further investigate the impact of MBS on weight loss-related asthma outcomes.

Supplementary Material

SUPPORTING INFORMATION

Additional supporting information can be found online in the Supporting Information section at the end of this article.

Abbreviations:

BMI

body mass index

BPD/DS

biliopancreatic diversion with duodenal switch

CENTRAL

Cochrane Central Register of Controlled Trials databases

LAGB

laparoscopic adjustable gastric banding

LSG

laparoscopic sleeve gastrectomy

MBS

metabolic and bariatric surgery

NIH

National Institutes of Health

NIHR

National Institute for Health Research

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

RYGB

Roux-en-Y gastric bypass

WHO

World Health Organization