Upper Gastrointestinal Studies After Laparoscopic Sleeve Gastrectomy: A Study that Prolongs Length of Stay

Abstract

Background and Objectives:

Laparoscopic sleeve gastrectomy (LSG) has become a common bariatric procedure performed in the United States. The purpose of our study was to determine factors affecting length of stay (LOS) after LSG in a University Hospital in the United States.

Methods:

One hundred eighty-seven patients underwent LSG over two 6-month periods in 2013 and 2014. Patients were selected by their surgeon to undergo a routine upper gastrointestinal (UGI) study (UGI group) or to proceed directly to a liquid diet (No UGI group). Primary endpoint was LOS. Secondary endpoint was surgical morbidity. One hundred thirty-four patients were in the UGI group and 53 patients were in the No UGI group.

Results:

Postoperative complications were 8% in the UGI group and 4% in the No UGI group (p = 0.28). There were no leaks and no significant difference in 90-day readmission rates (p = 0.44). LOS was 58.9 h in the UGI group and 51.4 h in the No UGI group (p = 0.009). Multivariate logistic regression analysis was used to identify independent predictors of an LOS >48 h. Performing a postoperative UGI study was the strongest predictor of an LOS >48 h with an odds ratio (OR) of 1.524 (p = 0.01) followed by increasing American Society of Anesthesiology (ASA) score (OR = 2.18, p = 0.03).

Conclusions:

Patients undergoing a postoperative UGI study are more likely to have an increased LOS than patients in the No UGI group with no difference in surgical morbidity, leak rate, or readmission rate. Eliminating routine UGI studies after LSG is an effective strategy to decrease LOS.

Introduction

Laparoscopic sleeve gastrectomy (LSG) has become one of the most common bariatric procedures performed in the United States.¹ As healthcare costs continue to increase and length of stay (LOS) continues to factor into reimbursement, it becomes important to recognize factors that contribute to increasing LOS.

Several previously described factors affecting LOS include the necessary medical comorbidities required for patients with a body mass index (BMI) of 35 kg/m² to be eligible for surgery, as well as the baseline lack of mobility of bariatric patients.^2–4 Moreover, intraoperative and postoperative complications such as bleed, respiratory issues, and especially leak are known risk factors to prolong LOS.

One potential variable that delays patients from initiating a liquid diet on postoperative day 1 is the upper gastrointestinal (UGI) series. In some institutions, this is a routine study that is performed before allowing the patient to consume any diet. The patient is asked to swallow gastrograffin and fluoroscopy is performed to confirm that there is no leak from the staple line and that contrast flows easily into the small bowel.

In this study, we sought to determine potential parameters that affect LOS after LSG. Importantly, we are evaluating the utility of routine UGI series in morbidly obese patients undergoing LSG. Our hypothesis is that routine UGI series are not necessary in morbidly obese patients undergoing LSG, which would translate into a shorter LOS.

Materials and Methods

Patients studied

The records of all patients undergoing LSG were reviewed using our database approved by the Institutional Review Board. Inclusion criteria were any morbidly obese patient who has chosen to undergo LSG from January 1, 2014, to June 30, 2015. Exclusion criteria included any patient who had previously been subjected to any type of bariatric surgery. All patients undergo preoperative UGI endoscopy. In total, 187 LSGs were performed by two surgeons (A.P. and G.D.) during the study period. The database was mined retrospectively. Before changing clinical practice from routine UGI to selective UGI in June 30, 2014, a total of 91 patients underwent routine UGI on postoperative day 1. After this change, patients underwent LSG and proceeded directly to a liquid diet (n = 96); selective UGI was only obtained at the surgeon's discretion at the end of the procedure (n = 42) either secondary to clinical concern for a leak (e.g., tachycardia and low urine output) or to assess the anatomy of the sleeve stomach in patients with persistent nausea or vomiting. This methodology for reporting results after implementing a clinical change in practice has been previously described.⁵

One hundred thirty-three patients were included in the UGI group and 54 patients were included in the No UGI group. The two groups were compared based on independent variables such as gender, age, BMI, American Society of Anesthesiology (ASA) score, and the following preoperative comorbidities: hypertension, diabetes mellitus, and hyperlipidemia. The primary endpoint of the study was LOS. The secondary endpoint was surgical morbidity. Postoperative complications were classified into various organ-specific categories. Genitourinary complications included urinary retention, requiring replacement of a urinary catheter. Hematological complications included postoperative bleeding, requiring blood transfusions or venous thromboembolism. Cardiac complications included arrhythmias and acute myocardial infarctions. Respiratory complications included acute respiratory distress. GI complications included food intolerance leading to clinically significant nausea and/or vomiting either delaying discharge (>3 days postoperatively) or causing readmission. Endocrine and infectious complications included hyperglycemia, requiring a continuous infusion of insulin and surgical site infection or urinary tract infection, respectively.

Statistics

Statistical analyses were performed using Graphpad Prism software version 5.03 (GraphPad Software, Inc., La Jolla, CA) and MYSTAT version 12 (SYSTAT Software, Inc., Chicago, IL). Categorical variables were compared using Fisher's exact test or chi-square test when appropriate, whereas continuous variables were compared using the Kruskal–Wallis test or Mann–Whitney U test (two-tailed). Univariate binary logistic regression analysis was used to identify independent predictors of patients with an LOS >48 h. Those parameters with a p < 0.15 as well as ASA, BMI, and the occurrence of a postoperative complication were included in the multivariable analysis. Multivariate logistic regression analysis was used to identify independent predictors of an LOS >48 h. Both forward and backward stepwise regression analyses were used for removing parameters with a p > 0.15. All results are expressed as mean ± SD, unless specified otherwise. The null hypothesis was rejected when α < 0.05.

Surgical technique

Our technique has been previously described.^6,7 In brief, the short gastric vessels are transected using an advanced surgical energy device and the hiatus is dissected. All hiatal hernias are repaired primarily. The stomach is stapled over a 40-French bougie. The first staple fire is unreinforced because of stomach thickness and this is oversewn for hemostasis. The rest of the staple line is completed using reinforced staple loads. A leak test using methylene blue dye is conducted. LSG is performed with the same technique by both surgeons at our institution.

Results

Patients in each group were not significantly different with respect to age, gender, BMI, ASA score, and comorbidity profiles (Table 1). Table 2 lists perioperative parameters and outcomes. Postoperative complications were 8% in the UGI group and 4% in the No UGI group (p = 0.28). The most common complication category in the UGI group was cardiovascular (n = 3). Two patients in the No UGI group experienced prolonged food intolerance leading to readmission. There were no leaks or deaths and no significant differences in 90-day readmission rates between the two groups (p = 0.44).

Table 1.

Preoperative Parameters

Parameter	Routine UGI (N = 134)	No UGI (N = 53)	p
Gender, male, n (%)	47 (36)	14 (26)	0.26
Age, years	42 ± 13	41 ± 12	0.44
BMI, kg/m²	46.4 ± 7.5	46.0 ± 6.6	0.80
ASA score, median (range)	3 (2–4)	3 (2–3)	0.57
Preoperative comorbidities, n (%)
HTN	65 (49)	22 (42)	0.39
DM2	28 (21)	6 (11)	0.13
HLD	44 (33)	13 (24)	0.27
OSA	54 (40)	14 (26)	0.08

ASA, American Society of Anesthesiology; BMI, body mass index; DM2, diabetes mellitus type 2; HLD, hyperlipidemia; HTN, hypertension; OSA: obstructive sleep apnea; UGI, upper gastrointestinal.

Table 2.

Perioperative Parameters and Outcomes

Parameter	Routine UGI (N = 134)	No UGI (N = 53)	p
Intraoperative complications	1/134 (1%)	0	0.53
Postoperative complications	11/134 (8%)	2/53 (4%)	0.28
Cardiovascular	3	0
Endocrine	1	0
Gastrointestinal	1	2
Genitourinary	1	0
Hematological	2	0
Infectious	2	0
Respiratory	1	0
Leak rate	0	0	1.0
Readmission rate (90 days)	9/134 (7%)	2/53 (4%)	0.44
Length of stay, h	58.9 ± 20.8	51.4 ± 10.8	0.009

The LOS was 58.9 h in the UGI group and 51.4 h in the No UGI group (p = 0.009) (Fig. 1). Routine use of UGI series was the only independent predictor of LOS >48 h on univariate analysis (p < 0.05). Multivariate logistic regression analysis was used to identify independent predictors of an LOS >48 h. Variables with a p < 0.15 were included as well as ASA, BMI, and occurrence of a postoperative complication. Performing a postoperative UGI study was the strongest predictor of an LOS more than 48 h in this cohort with an odds ratio (OR) of 1.524 (p = 0.01) on multivariate logistic regression analysis as listed in Table 3. The ASA score was the only other predictor of an LOS greater than 48 h on multivariate analysis with an OR of 2.18 (p = 0.03). Patient BMI was not an independent predictor of an increased LOS (p = 0.05) nor was obstructive sleep apnea (p = 0.09).

FIG. 1.

The box and whisker plots represent the 10th, 25th, 50th (median), 75th, and 90th percentiles. The x-axis represents the patients who underwent routine UGI series after surgery (UGI) and patients who did not undergo routine UGI series after surgery (No UGI), whereas the y-axis represents the length of stay in hours. UGI, upper gastrointestinal.

Table 3.

Multivariate Logistic Regression Analysis

Variable	Odds ratio	Ratio	p
UGI performed	1.524	1.087–2.132	0.01
ASA	2.183	1.068–4.444	0.03
BMI	0.995	0.913–1.000	0.05
OSA	0.752	0.542–1.045	0.09
Postoperative complication (yes)	1.621	0.851–3.086	0.14

Discussion

Routine use of UGI series after LSG prolonged LOS in our study, most likely because of delay in diet progression. Moreover, patients with higher ASA scores were also more likely to have longer hospitalizations.

The purpose of performing routine UGI series after LSG is to capture early technical complications, such as a leak. The leak rate after LSG varies in the literature and is cited to range between 1% and 2.4%.^8–12 The sensitivity of an UGI study to detect a leak varies in the literature. A meta-analysis that included 19 studies and a total of 10,139 patients who underwent bariatric surgery from 2003 to 2013 found that UGI has an overall sensitivity of 54% with a standard deviation of 36% in the reported studies.¹³ Studies like this have led some physicians to prefer to follow the patient clinically with tachycardia being a common indicator of a leak in addition to respiratory distress, fever, elevated white blood cell count, and physical examination findings. An UGI or even a CT scan can then potentially confirm the leak.¹⁴

Although LOS is not directly examined, the use of routine UGI series has been examined in the bariatric literature. White et al. demonstrated that UGI may be used selectively after laparoscopic Roux-en-y gastric bypass.⁵ Similarly, Wahby et al. studied a series of 712 consecutive LSG patients who underwent routine UGI series.¹⁵ The authors concluded that it is not needed to detect leakage unless clinically indicated in selected cases. Moreover, the authors determined that CT scan with oral contrast was the modality of choice. The study also advocated for use of the methylene blue test intraoperatively and detected a leak in 3.9% cases that they were able to immediately correct, reducing their postoperative leak rate to 1.4%. The study did not address LOS or what factors may contribute to it. Mittermair et al. studied 161 patients who underwent LSG with UGI series performed on postoperative day 1.¹⁶ The authors reported a 1.9% leak rate, prompting them to also conclude that clinical signs and symptoms such as tachycardia, pain, or fever are superior to routine imaging. Again, this study did not address any potential effects on LOS.

LOS studies have predominantly focused on various postoperative pathways that can be implemented before, during, and after surgery. Awad et al. implemented an enhanced recovery after bariatric surgery protocol.¹⁷ This protocol consists of early mobilization and feeding, avoidance of fluid overload, incentive spirometry, and the use of prokinetics and laxatives. They did not use routine UGI studies. Their mean LOS for LSG was 2.3 days. Others have taken it a step further and implemented same-day surgery for LSG in select patients by implementing a strict preoperative routine, carefully selected anesthetics, and postoperative pathways with strict criteria for readmission to clinic for hydration.¹⁸

Limitations of this study include a relatively small sample size with no intraoperative or postoperative leaks in either group, therefore, precluding our ability to calculate the sensitivity of the methylene blue test or an UGI study. The patients were not randomized to the UGI group or No UGI group but instead were relegated to each group based on clinician preference in the 2014 cohort. Nevertheless, this study does address the implications routine UGI series may have on LOS and thus offers clinicians an independent postoperative variable that can mitigate LOS.

Conclusion

In conclusion, eliminating routine UGI studies after LSG is an effective strategy to decrease LOS without increasing patient morbidity, mortality, or readmission rates. We recommend selective use of UGI series to evaluate morbidly obese patients undergoing LSG. A selective approach should be used in conjunction with clinical signs and symptoms suggestive of a leak.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Nguyen

, Nguyen

, Gebhart

, Hohmann

. Changes in the makeup of bariatric surgery: a national increase in use of laparoscopic sleeve gastrectomy. J Am Coll Surg, 2013; 216:252–257.

Froylich

, Corcelles

, Davis

, et al. Factors associated with length of stay in intensive care after bariatric surgery. Surg Obes Relat Dis, 2015; pii: S1550-7289(15)01084-9.

Calle

, Thun

, Petrelli

, Rodriguez

, Heath

Jr . Body-mass index and mortality in a prospective cohort of U.S. adults. N Engl J Med, 1999; 341:1097–1105.

Pieracci

, Barie

, Pomp

. Critical care of the bariatric patient. Crit Care Med, 2006; 34:1796–1804.

White

, Han

, Lewis

, et al. Selective approach to use of upper gastroesophageal imaging study after laparoscopic Roux-en-Y gastric bypass. Surg Obes Relat Dis, 2008; 4:122–125.

Afaneh

, Costa

, Pomp

, Dakin

. A prospective randomized controlled trial assessing the efficacy of omentopexy during laparoscopic sleeve gastrectomy in reducing postoperative gastrointestinal symptoms. Surg Endosc, 2015; 29:41–47.

Moy

, Pomp

, Dakin

, Parikh

, Gagner

. Laparoscopic sleeve gastrectomy for morbid obesity. Am J Surg, 2008; 196:e56–e59.

Weiner

, El-Sayes

, Theodoridou

, Weiner

, Scheffel

. Early post-operative complications: incidence, management, and impact on length of hospital stay. A retrospective comparison between laparoscopic gastric bypass and sleeve gastrectomy. Obes Surg, 2013; 23:2004–2012.

Sakran

, Goitein

, Raziel

, et al. Gastric leaks after sleeve gastrectomy: a multicenter experience with 2,834 patients. Surg Endosc, 2013; 27:240–245.

10.

Zellmer

, Mathiason

, Kallies

, Kothari

. Is laparoscopic sleeve gastrectomy a lower risk bariatric procedure compared with laparoscopic Roux-en-Y gastric bypass? A meta-analysis. Am J Surg, 2014; 208:903–910.

11.

Aurora

, Khaitan

, Saber

. Sleeve gastrectomy and the risk of leak: a systematic analysis of 4,888 patients. Surg Endosc, 2012; 26:1509–1515.

12.

Abelson

, Afaneh

, Dolan

, Chartrand

, Dakin

, Pomp

. Laparoscopic sleeve gastrectomy: co-morbidity profiles and intermediate-term outcomes. Obes Surg, 2015; DOI: 10.1007/s11695-015-2002-2.

13.

Mbadiwe

, Prevatt

, Duerinckx

, Cornwell

, Fullum

, Davis

. Assessing the value of routine upper gastrointestinal contrast studies following bariatric surgery: a systematic review and meta-analysis. Am J Surg, 2015; 209:616–622.

14.

Bingham

, Shawhan

, Parker

, Wigboldy

, Sohn

. Computed tomography scan versus upper gastrointestinal fluoroscopy for diagnosis of staple line leak following bariatric surgery. Am J Surg, 2015; 209:810–814.

15.

Wahby

, Salama

, Elezaby

, et al. Is routine postoperative gastrografin study needed after laparoscopic sleeve gastrectomy? Experience of 712 cases. Obes Surg, 2013; 23:1711–1717.

16.

Mittermair

, Sucher

, Perathoner

, Wykypiel

. Routine upper gastrointestinal swallow studies after laparoscopic sleeve gastrectomy are unnecessary. Am J Surg, 2014; 207:897–901.

17.

Awad

, Carter

, Purkayastha

, et al. Enhanced recovery after bariatric surgery (ERABS): clinical outcomes from a tertiary referral bariatric centre. Obes Surg, 2014; 24:753–758.

18.

Garofalo

, Denis

, Abouzahr

, Garneau

, Pescarus

, Atlas

. Fully ambulatory laparoscopic sleeve gastrectomy: 328 consecutive patients in a Single Tertiary Bariatric Center. Obes Surg, 2015; DOI: 10.1007/s11695-015-1984-0.