Dehydration risk factors and impact after bariatric surgery: an analysis using a national database

Journal Pre-proof Dehydration risk factors and impact after bariatric surgery: an analysis using a national database Tommy Ivanics, MD, Hassan Nasser, MD, Shravan Leonard-Murali, MD, Jeffrey Genaw, MD PII:

S1550-7289(19)30986-4

DOI:

https://doi.org/10.1016/j.soard.2019.09.054

Reference:

SOARD 3912

To appear in:

Surgery for Obesity and Related Diseases

Received Date: 9 June 2019 Revised Date:

2 August 2019

Accepted Date: 2 September 2019

Please cite this article as: Ivanics T, Nasser H, Leonard-Murali S, Genaw J, Dehydration risk factors and impact after bariatric surgery: an analysis using a national database, Surgery for Obesity and Related Diseases (2019), doi: https://doi.org/10.1016/j.soard.2019.09.054. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc. on behalf of American Society for Bariatric Surgery.

Title: Dehydration risk factors and impact after bariatric surgery: an analysis using a national database

Authors: Tommy Ivanics MDa, Hassan Nasser MDa, Shravan Leonard-Murali, MDa, Jeffrey Genaw, MDa Affiliations: aDepartment of Surgery, Henry Ford Hospital, Detroit, MI, United States

Corresponding author: Hassan Nasser, MD, [email protected] 2799 W. Grand Blvd. Department of Surgery CFP 127 Detroit, MI 48202-2689 313-916-3056 Phone 313-916-5811 Fax

Running head: Dehydration after bariatric surgery

Title: Dehydration risk factors and impact after bariatric surgery: an analysis using a national database Abstract Background: Dehydration is a common complication following bariatric surgery and 5

often quoted as the reason for emergency department (ED) visits and readmission. Objective: We sought to investigate risk factors for dehydration after bariatric surgery and evaluate its impact on ED visits and readmission. Setting: The Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) database.

10

Methods: We used the MBSAQIP database to identify patients who underwent laparoscopic sleeve gastrectomy (LSG) or laparoscopic Roux-en-Y gastric bypass (LRYGB) from 2016 through 2017. The primary outcome was need for outpatient treatment of dehydration within 30 days post-surgery. Secondary outcomes were association between need for outpatient dehydration therapy and 30-day readmission

15

or ED evaluation not resulting in admission. Results: Of 256,817 patients, 73% underwent LSG and 27% LRYGB. Of the 9,592 patients who required dehydration treatment, they were more often younger than age 40, female, black, had ≥ 3-day length of stay (LOS) during their index admission, and experienced a postoperative complication. More patients receiving LRYGB than LSG

20

required treatment for dehydration. On multivariable analysis, independent risk factors for postoperative dehydration treatment included LRYGB, LOS ≥ 3 days, gastroesophageal reflux disease, hypertension, previous deep vein thrombosis, chronic steroid/immunosuppression, and a postoperative complication. Patients who developed dehydration requiring treatment compared to those that did not had

1

25

adjusted odds ratio of 3.7 (95% CI 3.44-3.96; p<0.001) and 22 (95% CI 21.05-23.06; p<0.001) of readmission and ED visit. Conclusion: Dehydration is a strong risk factor for postoperative ED visits and readmission. Closer surveillance and proactive measures for those at higher risk may potentially prevent the development of postoperative dehydration.

30

35

40

45

2

Keywords: bariatric surgery; sleeve gastrectomy; gastric bypass; dehydration; 50

readmission; MBSAQIP

3

Background The use of bariatric surgery to treat morbid obesity has increased in recent years both in the United States and worldwide.(1) Emergency department (ED) visits and 55

readmission after bariatric surgery are important quality metrics, which may negatively impact bariatric centers’ accreditation, funding, and reimbursement, as well as lead to significantly increased health care costs and decreased patient satisfaction and quality of life. Reducing hospital readmissions has been associated with reduced Medicare spending and a means to improve the quality of patient care.(4–

60

8)

Common complications following bariatric surgery are nausea, vomiting, and dehydration.(9) The recognition of the importance of dehydration, a potentially preventable complication,(10) for potential readmission has led many bariatric centers to develop strategies to mitigate and prevent its development including thorough post65

discharge instructions, close postoperative surveillance, and ambulatory rehydration. It is not clear what patients are at risk for dehydration and what the national impact of requiring dehydration therapy after bariatric surgery has on ED visits and readmission rates. Using a national database, we aimed to identify risk factors for the

70

development of postoperative dehydration requiring treatment following bariatric surgery. We hypothesized that it is possible to identify a subset of patients with elevated risk for dehydration which may be helpful to guide targeted measures to improve outcomes

75

4

Materials and Methods Data source This retrospective study used the Metabolic and Bariatric Surgery Accreditation and 80

Quality Improvement Program (MBSAQIP) Participant Use Data File (PUF) which is a Health Insurance Portability and Accountability Act (HIPAA)-compliant data file.(11) This data set contains patient-level aggregate data with no identifiable information on hospital, health care provider or patients and is submitted by 832 MBSAQIP-participating centers in the United States and Canada. Data entry is

85

performed by certified clinical reviewers with specific training in metabolic and bariatric surgery. Quality of data is assessed periodically by MBSAQIP through data integrity audits.(11) A full set of variables with definitions included in the MBSAQIP PUF file are available in the PUF user guide.(11) Study population

90

This study was deemed exempt from Institutional Review Board review at our institution. We accessed the MBSAQIP database from January 1, 2016 to December 31, 2017 to identify patients who underwent laparoscopic sleeve gastrectomy (LSG), defined by current procedural terminology (CPT) code 43775, and laparoscopic Roux-en-Y gastric bypass (LRYGB), defined by CPT codes 43644 and 43645. These

95

two procedures were selected because they are the most commonly performed bariatric procedures. Patients undergoing emergent surgery, conversions or revisions, age less than 18 years, and without 30-day follow-up data were excluded. Thirty-day follow up data was missing for 12,471 out of 269,288 (4.63%). The remaining sample available for analysis included 256,817 patients. A flow diagram of inclusion and

100

exclusion criteria is depicted in Figure 1. Only procedures utilizing the conventional

5

laparoscopic approach were included and conversions to any other approach were excluded. Variables analyzed Covariates related to patient characteristics included age, sex, race, preoperative 105

body-mass index (BMI) closest to surgery, and American Society of Anesthesiologists (ASA) class. Relevant patient comorbidities included gastroesophageal reflux disease (GERD), coronary artery disease (CAD), hypertension (HTN), hyperlipidemia, diabetes mellitus, recent smoking history, chronic obstructive pulmonary disease, oxygen dependence, obstructive sleep apnea

110

(OSA), renal insufficiency, dialysis, deep vein thrombosis (DVT) or pulmonary embolism (PE), preoperative anticoagulation, inferior vena cava filter in place, chronic immunosuppression/steroid use, previous obesity/foregut surgery, use of an assistive device for ambulation, and limited functional status. Operative variables collected included the placement of a drain at the initial operation, LOS ≥ 3 days, and

115

occurrence of any postoperative complication. The presence of any complication was defined as the occurrence of any of the following: acute renal failure requiring dialysis, progressive renal insufficiency, cardiac arrest, myocardial infarction, unplanned admission to the intensive care unit (ICU), coma > 24 hours, cerebrovascular accident, ventilator > 48 hours, unplanned intubation, pneumonia,

120

surgical site infection (superficial, deep/incisional, and organ space), sepsis, septic shock, urinary tract infection, DVT, PE, anticoagulation for presumed DVT or PE, bleeding requiring a transfusion, any intervention within 30 days, drain present at 30 days, and wound disruption. These outcomes were available in the database as binary variables. Patients with complications present at the time of surgery were excluded

125

from the analysis.

6

Outcomes The primary outcome measure was outpatient treatment for dehydration (nausea and vomiting, fluid, electrolyte, or Nutritional depletion) within 30 days postoperatively. This variable is intended to capture the number of treatments a patient received within 130

the 30-day postoperative timeframe for nausea and vomiting, fluid, electrolyte or nutritional depletion as an outpatient. Secondary outcomes were the association of the primary outcome with 30-day readmission or ED evaluation not resulting in an inpatient admission. All these variables were collected in the database as binary variables.

135

Statistical analysis Baseline variables and dehydration rates were reported using descriptive statistics. Univariate analysis used Pearson X2 test or Fischer’s exact test for categorical variables. To identify independent risk factors for the need for dehydration therapy, multivariable logistic regression was used to adjust for demographic, comorbidity,

140

and operative variables. Specifically, these risk factors included age, sex, race, BMI, GERD, Coronary artery disease, hypertension, hyperlipidemia, diabetes mellitus, smoking, COPD, oxygen dependence, obstructive sleep apnea, renal insufficiency, dialysis dependence, history of deep vein thrombosis, history of pulmonary embolism, anticoagulation status, IVC filter, presence of venous stasis, chronic

145

steroids/immunosuppression, previous obesity/foregut surgery, functional status, ASA class, operative time, abdominal drainage, length of stay greater than or equal to 3 days, presence of a postoperative complication. A purposeful selection algorithm was used with any variable with p<0.100 included in the model. Models addressed 3 patient groups: all patients, LSG, and LRYGB. The Hosmer-Lemeshow goodness-of-

150

fit test was used to evaluate goodness of fit of the model. Multivariate logistic

7

regression was also used to calculate the adjusted odds ratio (AOR) for readmission and for ED evaluation in patients requiring dehydration therapy. Statistical significance was set to p<0.05. All analysis was performed using IBM SPSS version 25. 155 Results Of 256,817 patients identified, 73% underwent LSG and 27% LRYGB. Overall and for each of the procedures, the majority of patients were female, white, functionally independent, with a preoperative BMI between 40-49 and ASA class 3 (Table 1). The 160

most common comorbidities for all patients included GERD (30.9%) and OSA (38.7%). Few patients had a LOS ≥ 3 days (9.85%) and experienced a postoperative complication (3.67%) (Table 1). Patients who required treatment for dehydration versus those who did not were more often younger than age 40 (18-29 years: 16.5% vs. 11.4% and 30-39 years:

165

31.9% vs. 25.4%; p<0.001), female (88.8% vs. 79.2%; p<0.001), black (23.9% vs. 17.3%; p<0.001), had a LOS ≥ 3 days during their index admission (16.2% vs. 9.60%; p<0.001), and experienced a postoperative complication (11.7% vs. 3.36%; p<0.001) (Table 2). No differences were noted between the groups for renal insufficiency, dialysis dependence, OSA, or CAD (Table 2).

170

More patients who underwent LRYGB vs. LSG required treatment for dehydration (4.8% vs. 3.3%). After adjustment on multivariable analysis, LRYGB had an AOR of 1.26 (95% confidence interval [CI] 1.20-1.32; p<0.001) for requiring postoperative dehydration treatment compared to LSG. Younger age, female gender, non-Hispanic race were risk factors for need for postoperative dehydration treatment.

175

Additional risk factors included LOS ≥ 3 days (AOR 1.36, 95% CI 1.28-1.44;

8

p<0.001), GERD (AOR 1.45, 95% CI 1.39-1.52; p<0.001), HTN (AOR 1.08, 95% CI 1.03-1.13; p=0.002), previous DVT (AOR 1.28, 95% CI 1.10-1.49; p=0.002), chronic steroid/immunosuppression (AOR 1.20, 95% CI 1.04-1.38; p=0.013), operative time (with every 60 minute increase) (AOR 1.06 (95% CI 1.03-1.09) and a postoperative 180

complication during the index hospital admission (AOR 3.18, 95% CI 2.96-3.41; p<0.001).Preoperative BMI did not represent an independent risk factor (Table 3). Of patients who required dehydration therapy (n=9592), 1421 (14.8%) were readmitted and 5287 (55.1%) presented to the ED. After adjustment, patients with requiring dehydration therapy had an AOR of 3.69 (95% CI 3.44-3.96; p<0.001) for

185

readmission and AOR of 22.03 (95% CI 21.05-23.06; p<0.001) for an ED visit.

Discussion This analysis of prospectively collected clinical data from the MBSAQIP database identified several risk factors for need for dehydration treatment following bariatric 190

surgery: LRYGB procedure, LOS ≥ 3 days, age younger than age 40, female sex, black race, GERD, HTN, previous DVT, prolonged operative time, and chronic steroid/immunosuppression. The strongest independent risk factor for requiring postoperative dehydration treatment was a postoperative complication during the index hospital admission. Patients who required dehydration treatment

195

postoperatively have 3.7 and 22.0 greater odds of being readmitted or presenting to the ED, respectively compared to patients who did not require dehydration treatment. Patients undergoing bariatric surgery who have several of these risk factors may warrant closer postoperative surveillance, such as more frequent postoperative office visits, follow-up phone calls and education. Additional education may be beneficial to

200

providers in regard to diuretic and antihypertensive therapy management

9

postoperatively for high-risk patients. This may include ensuring availability of postoperative antiemetic therapy and emphasis on recognition of signs and symptoms of dehydration. The constellation of nausea, vomiting, and dehydration, most common after bariatric surgery,(9,12,13) is typically multifactorial in etiology and may 205

include dysfunctional eating habits such as overeating.(14,15) Dehydration typically results due to decreased fluid intake, which may be exacerbated by watery stool and vomiting. Patients with obesity require a higher amount of maintenance fluids and thus are more prone to develop dehydration.(14) ED visits and readmissions, important quality metrics in bariatric surgery,

210

result in increased health care costs and decreased patient satisfaction and quality of life postoperatively.(16) Recently, the validity of using hospital readmission as a quality metric has been questioned.(17,18) Nevertheless, it remains an important recorded postoperative variable, which can be used to identify deficiencies in postoperative care and improve patient outcomes.

215

Several studies have investigated factors associated with readmissions following bariatric surgery.(7,9,13,19–24) These include technical factors, longer LOS, surgeon volume, hospital volume, insurance, socioeconomic level, and functional status. A recent study suggested that nearly half of readmissions are potentially preventable.(25) Daigle et al. identified bleeding and leak as the complication with the

220

largest effect on morbidity, reoperation, and ICU admission.(26) The same group proposed that targeting venous thromboembolism (VTE) would have the highest potential to reduce mortality and readmission rates. In our analysis, DVT and PE occurred more in the dehydration treatment group, and previous DVT was an independent risk factor for readmission postoperatively. The highest readmission

225

among bariatric procedures has been found to be in patients undergoing RYGB.(7) In

10

contrast to this study, Sippey at al. found that nausea, vomiting and dehydration were more commonly a reason for readmission in LSG than LRYGB. However, the focus of this study was to evaluate the outcomes of patients who received treatment for dehydration in an outpatient setting. It may also be related to a different studied time230

period (2012-2013). Changes in practice patterns and postoperative surveillance may have contributed to this difference as well. In a single-institution retrospective analysis of consecutive laparoscopic and open RYGB cases (n=1222), Kellogg et al. found nausea, vomiting, and dehydration to be the most common admitting diagnosis (26%) for patients with ED visits, readmissions, or reoperation within 90 days

235

postoperatively.(9-11) Our analysis similarly found that more patients who underwent LRYGB experienced dehydration and showed higher odds for readmission. With identification of risk factors for dehydration after bariatric surgery, strategies can focus on high-risk patients to help prevent or reduce this complication as well as associated ED visits and readmissions.

240

This study was limited by the retrospective nature of the data within MBSAQIP and subject to potential biases inherent to any retrospective analysis of multi-institutional datasets. The number of missing variables was low (4.63%) however the validity of data variables available in the dataset is dependent on data entry by certified clinical reviewers. Data quality is periodically evaluated by

245

MBSAQIP through data integrity audits to ensure its accuracy. Variations in coding between institutions cannot be entirely excluded as potential sources of bias. Parameters not included in the MBSAQIP dataset cannot be accounted or adjusted for. Data entry is performed by certified clinical reviewers with specific training in metabolic and bariatric surgery. Quality of data is assessed periodically by MBSAQIP

250

through data integrity audits.(11) The database contains only 30-day postoperative data

11

and thus admissions after that time are not captured. Details on when an ED visit or readmission occurred are also not captured. Because patients may have presented to another facility, actual rates may in fact be higher than what is currently reported in the database. Lastly, there is no information what type of treatment a patient received 255

as part of their outpatient dehydration treatment.

Conclusion Dehydration is a strong risk factor for postoperative ED visits and readmission after LSG or LRYGB. Patients at higher risk for requiring dehydration treatment include 260

those undergoing LRYGB, LOS ≥ 3 days, younger, female, black, experience a postoperative complication, and have GERD, HTN, previous DVT, and chronic steroids/immunosuppression. Closer surveillance and proactive measures for patients with elevated risk may serve to mitigate and potentially prevent the development of postoperative dehydration and improve outcomes.

265 Disclosures The authors have no conflicts of interests or source of funding to disclose.

270

275

12

References 1.

Lodhia NA, Morton JM. Weighing in on bariatric surgery: who and when? Int J Obes Suppl. 2012;2(S1):S47-S50. doi:10.1038/ijosup.2012.12

2. 280

Santry HP, Gillen DL, Lauderdale DS. Trends in bariatric surgical procedures. J Am Med Assoc. 2005. doi:10.1001/jama.294.15.1909

3.

Kizy S, Jahansouz C, Downey MC, Hevelone N, Ikramuddin S, Leslie D. National Trends in Bariatric Surgery 2012–2015: Demographics, Procedure Selection, Readmissions, and Cost. Obes Surg. 2017;27(11):2933-2939. doi:10.1007/s11695-017-2719-1

285

4.

Stone J, Hoffman G. Medicare hospital readmissions: issues, policy options and PPACP. CRS Rep Congr. 2010:11.

5.

Lawson EH, Hall BL, Louie R, et al. Association Between Occurrence of a Postoperative Complication and Readmission. Ann Surg. 2013;258(1):10-18. doi:10.1097/sla.0b013e31828e3ac3

290

6.

Postel M, Frank PN, Barry T, Satou N, Shemin R, Benharash P. The cost of preventing readmissions: Why surgeons should lead the effort. Am Surg. 2014.

7.

Saunders J, Ballantyne GH, Belsley S, et al. One-year readmission rates at a high volume bariatric surgery center: Laparoscopic adjustable gastric banding, laparoscopic gastric bypass, and vertical banded gastroplasty-Roux-en-Y

295

gastric bypass. Obes Surg. 2008;18(10):1233-1240. doi:10.1007/s11695-0089517-8 8.

Patterson WL, Peoples BD, Gesten FC. Predicting potentially preventable hospital readmissions following bariatric surgery. Surg Obes Relat Dis. 2015;11(4):866-872. doi:10.1016/j.soard.2014.12.019

300

9.

Kellogg TA, Swan T, Leslie DA, Buchwald H, Ikramuddin S. Patterns of

13

readmission and reoperation within 90 days after Roux-en-Y gastric bypass. Surg Obes Relat Dis. 2009;5(4):416-423. doi:10.1016/j.soard.2009.01.008 10.

Chen J, Mackenzie J, Zhai Y, et al. Preventing Returns to the Emergency Department FollowingBariatric Surgery. Obes Surg. 2017;27(8):1986-1992.

305

doi:10.1007/s11695-017-2624-7 11.

MBSAQIP: User Guide for the 2017 Participant Use Data File (PUF). 2018;(October):1-20. https://www.facs.org/qualityprograms/mbsaqip/participant-use.

12. 310

Sippey M, Kasten KR, Chapman WHH, Pories WJ, Spaniolas K. 30-day readmissions after sleeve gastrectomy versus Roux-en-Y gastric bypass. Surg Obes Relat Dis. 2016;12(5):991-996. doi:10.1016/j.soard.2016.01.036

13.

Aman MW, Stem M, Schweitzer MA, Magnuson TH, Lidor AO. Early hospital readmission after bariatric surgery. Surg Endosc. 2016;30(6):2231-2238. doi:10.1007/s00464-015-4483-4

315

14.

Shikora SA, Kim JJ, Tarnoff ME. Nutrition and gastrointestinal complications of bariatric surgery. Nutr Clin Pract. 2007;22(1):29-40. doi:10.1177/011542650702200129

15.

Ziegler O, Sirveaux MA, Brunaud L, Reibel N, Quilliot D. Medical follow up after bariatric surgery: nutritional and drug issues General recommendations

320

for the prevention and treatment of nutritional deficiencies. Diabetes Metab. 2009;35(6 PART II):544-557. doi:10.1016/S1262-3636(09)73464-0 16.

Encinosa WE, Bernard DM, Chen CC, Steiner CA. Healthcare utilization and outcomes after bariatric surgery. Med Care. 2006;44(8):706-712. doi:10.1097/01.mlr.0000220833.89050.ed

325

17.

Parina RP, Chang DC, Rose JA, Talamini MA. Is a low readmission rate

14

indicative of a good hospital? J Am Coll Surg. 2015;220(2):169-176. doi:10.1016/j.jamcollsurg.2014.10.020 18.

Merkow RP, Ju MH, Chung JW, et al. Underlying reasons associated with hospital readmission following surgery in the United States. JAMA - J Am Med

330

Assoc. 2015;313(5):483-495. doi:10.1001/jama.2014.18614 19.

Sethi M, Patel K, Zagzag J, et al. Thirty-Day Readmission After Laparoscopic Sleeve Gastrectomy—a Predictable Event? J Gastrointest Surg. 2016;20(2):244-252. doi:10.1007/s11605-015-2978-x

20. 335

Willson TD, Gomberawalla A, Mahoney K, Lutfi RE. Factors Influencing 30Day Emergency Visits and Readmissions after Sleeve Gastrectomy: Results from a Community Bariatric Center. Obes Surg. 2015;25(6):975-981. doi:10.1007/s11695-014-1546-x

21.

Weller WE, Rosati C, Hannan EL. Relationship Between Surgeon and Hospital Volume and Readmission after Bariatric Operation. J Am Coll Surg.

340

2007;204(3):383-391. doi:10.1016/j.jamcollsurg.2006.12.031 22.

Macht R, George J, Ameli O, Hess D, Cabral H, Kazis L. Factors associated with bariatric postoperative emergency department visits. Surg Obes Relat Dis. 2016;12(10):1826-1831. doi:10.1016/j.soard.2016.02.038

23. 345

Celio AC, Kasten KR, Burruss MB, Pories WJ, Spaniolas K. Surgeon case volume and readmissions after laparoscopic Roux-en-Y gastric bypass: more is less. Surg Endosc. 2017;31(3):1402-1406. doi:10.1007/s00464-016-5128-y

24.

Saunders JK, Ballantyne GH, Belsley S, et al. 30-Day readmission rates at a high volume bariatric surgery center: Laparoscopic adjustable gastric banding, laparoscopic gastric bypass, and vertical banded gastroplasty-Roux-en-Y

350

gastric bypass. Obes Surg. 2007;17(9):1171-1177.

15

http://www.embase.com/search/results?subaction=viewrecord&from=export&i d=L351321334%5Cnhttp://dx.doi.org/10.1007/s11695-007-92103%5Cnhttp://sfx.library.uu.nl/utrecht?sid=EMBASE&issn=09608923&id=doi: 10.1007%2Fs11695-007-9210-3&atitle=30-Day+readmission+rates+at. 355

25.

Petrick A, Brindle SA, Vogels E, Dove J, Parker D, Gabrielsen J. The readmission contradiction: toward clarifying common misconceptions about bariatric readmissions and quality improvement. Surg Obes Relat Dis. 2018;14(7):1026-1032. doi:10.1016/j.soard.2018.02.017

26. 360

Daigle CR, Brethauer SA, Tu C, et al. Which postoperative complications matter most after bariatric surgery? Prioritizing quality improvement efforts to improve national outcomes. Surg Obes Relat Dis. 2018;14(5):652-657. doi:10.1016/j.soard.2018.01.008

365

370

375

16

Figure Legend Figure 1. Study population flow chart

17

Table 1. Patient characteristics, co-morbidities, and operative variables by surgery type. Demographics Age, years 18-29 30-39 40-49 50-59 ≥60 Sex Male Female Race White Black Other Unknown/Not reported Hispanic Yes No Unknown BMI, kg/m2 <35 35-39 40-49 50-59 ≥60 Co-morbidities GERD Coronary artery disease Hypertension Hyperlipidemia Diabetes mellitus Smoking COPD Oxygen dependent Obstructive sleep apnea Renal insufficiency Dialysis Deep vein thrombosis Pulmonary embolism Anticoagulation IVC filter Venous stasis Chronic steroid/immunosuppression Previous obesity/foregut surgery Functional status Independent Partially Dependent Totally Dependent Limited ambulation ASA class 1-2 3 4-5 Operative variables Operative time, minutes Abdominal drainage LOS ≥ 3 days Postoperative complication

All patients N = 256817

LSG N = 187412

LRYGB N = 69405

29812 (11.6%) 65960 (25.7%) 74702 (29.1%) 57087 (22.2%) 29256 (11.4%)

22683 (12.1%) 48927 (26.1%) 54341 (29.0%) 40512 (21.6%) 20949 (11.2%)

7129 (10.3%) 17033 (24.5%) 20361 (29.3%) 16575 (23.9%) 8307 (12.0%)

52429 (20.4%) 204388 (79.6%)

38899 (20.8%) 148513 (79.2%)

13530 (19.5%) 55875 (80.5%)

187072 (72.8%) 45132 (17.6%) 3037 (1.2%) 21576 (8.4%)

135050 (72.1%) 35711 (19.0%) 1979 (1.1%) 14672 (7.8%)

52022 (75.0%) 9421 (13.5%) 1058 (1.5%) 6904 (10.0%)

32439 (12.6%) 200261 (78.0%) 24117 (9.4%)

23645 (12.6%) 146280 (78.1%) 17487 (9.3%)

8794 (12.7%) 53981 (77.8%) 6630 (9.6%)

8645 (3.4%) 57865 (22.5%) 132703 (51.7%) 45027 (17.5%) 12577 (4.9%)

6682 (3.6%) 44150 (23.6%) 97039 (51.8%) 30886 (16.5%) 8655 (4.6%)

1963 (2.8%) 13715 (19.8%) 35664 (51.4%) 14141 (20.4%) 3922 (5.7%)

79470 (30.9%) 8149 (3.17%) 123748 (48.2%) 61143 (23.8%) 67457 (26.3%) 21658 (8.43%) 4286 (1.67%) 1817 (0.71%) 99357 (38.7%) 1581 (0.62%) 744 (0.29%) 4080 (1.59%) 3065 (1.19%) 7018 (2.73%) 1588 (0.62%) 2786 (1.08%) 4561 (1.78%) 3814 (1.49%)

52557 (28.0%) 5696 (3.04%) 87441 (46.7%) 41307 (22.0%) 43155 (23.0%) 16063 (8.57%) 3019 (1.61%) 1181 (0.63%) 68696 (36.7%) 1142 (0.61%) 637 (0.34%) 2765 (1.48%) 2142 (1.14%) 5035 (2.69%) 1147 (0.61%) 1894 (1.01%) 3448 (1.84%) 2662 (1.42%)

26913 (38.8%) 2453 (3.53%) 36307 (52.3%) 19836 (28.6%) 24302 (35.0%) 5595 (8.06%) 1267 (1.83%) 636 (0.92%) 30661 (44.2%) 439 (0.63%) 107 (0.15%) 1315 (1.89%) 923 (1.33%) 1983 (2.86%) 441 (0.64%) 892 (1.29%) 1113 (1.60%) 1152 (1.66%)

254312 (99.0%) 1556 (0.6%) 949 (0.4%) 3726 (1.45%)

185593 (99.0%) 1063 (0.6%) 756 (0.4%) 2543 (1.36%)

68719 (99.0%) 493 (0.7%) 193 (0.3%) 1183 (1.70%)

58061 (22.6%) 190127 (74.0%) 8629 (3.4%)

46298 (24.7%) 135521 (72.3%) 5593 (3.0%)

11763 (16.9%) 54606 (78.7%) 3036 (4.4%)

82.6 (45.4) 47053 (18.3%) 25289 (9.85%) 9430 (3.67%)

69.9 (34.8) 28221 (15.1%) 14452 (7.71%) 5045 (2.69%)

116.9 (52.3) 18832 (27.1%) 10837 (15.6%) 4385 (6.32%)

p-value

< 0.001

< 0.001

< 0.001

0.195

< 0.001

< 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.505 < 0.001 < 0.001 < 0.001 0.019 0.502 < 0.001 < 0.001 < 0.001 < 0.001

< 0.001 < 0.001

< 0.001 < 0.001 < 0.001 < 0.001

LSG laparoscopic sleeve gastrectomy, LRYGB laparoscopic Roux-en-Y gastric bypass, BMI body mass index, GERD gastroesophageal reflux disease, COPD chronic obstructive pulmonary disease, IVC inferior vena cava, ASA American Society of Anesthesiologists, LOS length of stay Frequency (%) for categorical variables Mean (standard deviation) for continuous variables

Table 2. Patient and operative characteristics by dehydration treatment requirement. Demographics Age, years 18-29 30-39 40-49 50-59 ≥60 Sex Male Female Race White Black Other Unknown/Not reported Hispanic Yes No Unknown BMI, kg/m2 <35 35-39 40-49 50-59 ≥60 Co-morbidities GERD Coronary artery disease Hypertension Hyperlipidemia Diabetes mellitus Smoking COPD Oxygen dependent Obstructive sleep apnea Renal insufficiency Dialysis Deep vein thrombosis Pulmonary embolism Anticoagulation IVC filter Venous stasis Chronic steroid/immunosuppression Previous obesity/foregut surgery Functional status Independent Partially Dependent Totally Dependent Limited ambulation ASA class 1-2 3 4-5 Operative variables Operative time, minutes Abdominal drainage LOS ≥ 3 days Postoperative complication Procedure performed LSG LRYGB

Dehydration treatment N = 9592

No dehydration treatment N = 247225

p-value

1578 (16.5%) 3056 (31.9%) 2605 (27.2%) 1654 (17.2%) 699 (7.3%)

28234 (11.4%) 62904 (25.4%) 72097 (29.1%) 55433 (22.4%) 28557 (11.6%)

1071 (11.2%) 8521 (88.8%)

51358 (20.8%) 195867 (79.2%)

6657 (69.4%) 2294 (23.9%) 108 (1.1%) 533 (5.6%)

180415 (73.0%) 42838 (17.3%) 2929 (1.2%) 21043 (8.5%)

1049 (10.9%) 7968 (83.1%) 575 (6.0%)

31390 (12.7%) 192293 (77.8%) 23542 (9.5%)

265 (2.8%) 1957 (20.4%) 5040 (52.5%) 1800 (18.8%) 530 (5.5%)

8380 (3.4%) 55908 (22.6%) 127663 (51.6%) 43227 (17.5%) 12047 (4.9%)

3740 (39.0%) 281 (2.93%) 4394 (45.8%) 1971 (20.6%) 2377 (24.8%) 882 (9.20%) 183 (1.91%)

75730 (30.6%) 7868 (3.18%) 119354 (48.3%) 59172 (23.9%) 65080 (26.3%) 20776 (8.40%) 4103 (1.66%)

< 0.001 0.165 < 0.001 < 0.001 0.001 0.006 0.063

98 (1.02%) 3645 (38.0%) 62 (0.65%) 28 (0.29%) 227 (2.37%) 176 (1.83%) 299 (3.12%) 66 (0.69%) 106 (1.11%) 218 (2.27%) 120 (1.25%)

1719 (0.70%) 95712 (38.7%) 1519 (0.61%) 716 (0.29%) 3853 (1.56%) 2889 (1.17%) 6719 (2.72%) 1522 (0.62%) 2680 (1.08%) 4343 (1.76%) 3694 (1.49%)

< 0.001 0.159 0.695 0.967 < 0.001 < 0.001 0.019 0.375 0.845 < 0.001 0.053 0.025

9501 (99.1%) 69 (0.7%) 22 (0.2%) 147 (1.53%)

244811 (99.0%) 1487 (0.6%) 927 (0.4%) 3579 (1.45%)

2057 (21.4%) 7215 (75.2%) 320 (3.3%)

56004 (22.7%) 182912 (74.0%) 8309 (3.3%)

88.5 (50.7) 1781 (18.6%) 1553 (16.2%) 1117 (11.7%)

82.3 (45.1) 45272 (18.3%) 23736 (9.60%) 8313 (3.36%)

< 0.001 0.526 < 0.001 < 0.001

6257 (65.2%) 3335 (34.8%)

181155 (73.3%) 66070 (26.7%)

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

< 0.001

0.495 0.020

BMI body mass index, GERD gastroesophageal reflux disease, COPD chronic obstructive pulmonary disease, IVC inferior vena cava, ASA American Society of Anesthesiologists, LOS length of stay, LSG laparoscopic sleeve gastrectomy, LRYGB laparoscopic Roux-en-Y gastric bypass, Frequency (%) for categorical variables Mean (standard deviation) for continuous variables

Table 3. Adjusted odds ratio for factors associated with postoperative dehydration treatment. All patients N = 256817 AOR (95% CI) Procedure performed LSG LRYGB Demographics Age, years 18-29 30-39 40-49 50-59 ≥60 Sex Male Female Race White Black Other Unknown/Not reported Hispanic No Yes Unknown Co-morbidities BMI, kg/m2 <35 35-39 40-49 50-59 ≥60 GERD Coronary artery disease Hypertension Hyperlipidemia Diabetes mellitus Smoking COPD Oxygen dependent Obstructive sleep apnea Renal insufficiency Dialysis Deep vein thrombosis Pulmonary embolism Anticoagulation IVC filter Venous stasis Chronic steroid/immunosuppression Previous obesity/foregut surgery Functional status Independent Partially Dependent Totally Dependent Limited ambulation ASA class 1-2 3 4-5 Operative variables Operative time, every 60 minutes increase Abdominal drainage LOS ≥ 3 days Postoperative complication

LSG N = 187412 AOR (95% CI)

LRYGB N = 69405 AOR (95% CI)

1.00 0.81 (0.76 – 0.86) 0.57 (0.53 – 0.61) 0.44 (0.41 – 0.48) 0.35 (0.32 – 0.39)

1.00 0.79 (0.74 – 0.86) 0.58 (0.53 – 0.63) 0.42 (0.38 – 0.47) 0.33 (0.29 – 0.37)

1.00 0.83 (0.74 – 0.93) 0.55 (0.48 – 0.62) 0.46 (0.41 – 0.53) 0.38 (0.32 – 0.45)

0.52 (0.49 – 0.56) 1.00

0.50 (0.46 – 0.55) 1.00

0.53 (0.48 – 0.60) 1.00

1.00 1.33 (1.26 – 1.40) 0.95 (1.78 – 1.16) 0.80 (0.72 – 0.88)

1.00 1.30 (1.22 – 1.38) 0.77 (0.58 – 1.02) 0.82 (0.72 – 0.92)

1.00 1.40 (1.28 – 1.54) 1.21 (0.92 – 1.59) 0.78 (0.66 – 0.93)

1.00 0.84 (0.79 – 0.90) 0.69 (0.63 – 0.76)

1.00 0.81 (0.75 – 0.89) 0.74 (0.66 – 0.82)

1.00 0.89 (0.79 – 1.00) 0.62 (0.51 – 0.74)

1.00 1.02 (0.90 – 1.17) 1.04 (0.92 – 1.19) 1.01 (0.88 – 1.16) 0.97 (0.83 – 1.13) 1.45 (1.39 – 1.52)

1.00 1.00 (0.85 – 1.17) 1.02 (0.88 – 1.19) 1.00 (0.85 – 1.17) 0.93 (0.77 – 1.12) 1.45 (1.37 – 1.53)

1.00 1.09 (0.85 – 1.39) 1.10 (0.87 – 1.39) 1.04 (0.82 – 1.33) 1.04 (0.80 – 1.36) 1.45 (1.35 – 1.56)

1.08 (1.03 – 1.13) 0.99 (0.93 – 1.05) 0.97 (0.92 – 1.02) 1.01 (0.94 – 1.09) 1.13 (0.96 – 1.32) 1.28 (1.03 – 1.60)

1.08 (1.02 – 1.15) 0.97 (0.90 – 1.04) 0.99 (0.92 – 1.05) 0.99 (0.92 – 1.09) 1.18 (0.96 – 1.44) 1.22 (0.90 – 1.64) 1.10 (1.03 – 1.16)

1.07 (0.99 – 1.16) 1.02 (0.93 – 1.13) 0.94 (0.86 – 1.02) 1.03 (0.91 – 1.16)

1.28 (1.10 – 1.49) 1.20 (1.01 – 1.42) 1.13 (0.99 – 1.29)

1.38 (1.14 – 1.68) 1.08 (0.87 – 1.35) 1.15 (0.98 – 1.36)

1.12 (0.87 – 1.44) 1.41 (1.07 – 1.84)

1.20 (1.04 – 1.38) 0.81 (0.67 – 0.97)

1.31 (1.11 – 1.54) 0.74 (0.58 – 0.96)

1.00 1.26 (1.20 – 1.32)

1.00 1.03 (0.81 – 1.33) 0.65 (0.42 – 0.99)

1.00 1.10 (0.75 – 1.61) 0.47 (0.17 – 1.27)

1.00 1.10 (1.05 – 1.16) 1.08 (0.95 – 1.23)

1.00 1.13 (1.06 – 1.20) 1.08 (0.91 – 1.27)

1.06 (1.03 – 1.09)

1.03 (0.98 – 1.07) 0.90 (0.84 – 0.97) 1.48 (1.36 – 1.60) 3.49 (3.17 – 3.85)

1.36 (1.28 – 1.44) 3.18 (2.96 – 3.41)

1.37 (1.00 – 1.89)

1.08 (1.04 – 1.13) 1.22 (1.12 – 1.34) 2.88 (2.60 – 3.19)

LSG laparoscopic sleeve gastrectomy, LRYGB laparoscopic Roux-en-Y gastric bypass, AOR adjusted odds ratio, BMI body mass index, GERD gastroesophageal reflux disease, COPD chronic obstructive pulmonary disease, IVC inferior vena cava, ASA American Society of Anesthesiologists, LOS length of stay Significant variables are in bold defined as p<0.05

Highlights •

A postoperative complication was the strongest factor to need dehydration treatment

•

More patients undergoing LRYGB than LSG require treatment for dehydration

•

Dehydration is a risk factor for ED visits and readmission after bariatric surgery