- Email: [email protected]
Comparative Analysis of Perioperative Outcomes and Costs Between Laparoscopic and Open Antireflux Surgery
Comparative Analysis of Perioperative Outcomes and Costs Between Laparoscopic and Open Antireflux Surgery Francisco Schlottmann,
MD,
Paula D Strassle,
MSPH,
Marco G Patti,
MD, FACS
Laparoscopic antireflux surgery (LARS) has proven to be as effective as open antireflux surgery (OARS), but it is associated with a shorter hospital stay and a faster recover. The aims of this study were to assess the national use of LARS in the US and to compare the perioperative outcomes between laparoscopic and open antireflux procedures in a national cohort. STUDY DESIGN: A retrospective population-based analysis was performed using the National Inpatient Sample for the period 2000 to 2013. The study included adult patients (18 years and older) diagnosed with gastroesophageal reflux disease (GERD), who underwent either laparoscopic or open fundoplication. Multivariable linear and logistic regression, adjusted for patient demographics, comorbidities, and hospital characteristics were used to assess the effect of the laparoscopic approach on patient outcomes. RESULTS: A total of 75,544 patients were included, with 44,089 having LARS (58.4%) and 31,455 having OARS (41.6%). The rate of laparoscopic procedures increased from 24.8 LARS per 100 procedures in 2000, to 84.3 LARS per 100 procedures in 2013 (p < 0.0001). Patients undergoing laparoscopic surgery were less likely to experience postoperative venous thromboembolism, wound complications, infection, esophageal perforation, bleeding, cardiac failure, renal failure, respiratory failure, shock, and inpatient mortality. On average, the laparoscopic approach reduced length of stay by 2.1 days, and decreased hospital charges by $9,530. CONCLUSIONS: The use of the laparoscopic approach for the surgical treatment of GERD has increased significantly in the last decade in the US. This approach is associated with lower morbidity and mortality, shorter hospital stay, and lower costs for the health care system. (J Am Coll Surg 2017;-:1e7. 2016 by the American College of Surgeons. Published by Elsevier Inc. All rights reserved.)
BACKGROUND:
intervention because they have only partial control of symptoms, do not want to be on long-term medical treatment, or suffer complications related to proton pump inhibitor therapy. The most commonly performed antireflux operation is the Nissen fundoplication (360 degrees), which has long-term success in about 80% to 90% of patients.2-4 Laparoscopic antireflux surgery (LARS) was first reported in 19915,6 and has since become widely accepted because of the clear advantages of a minimally invasive approach (decreased pain, shorter hospital stay, faster recovery, and almost complete absence of incisional hernia), and because it is as effective as open antireflux surgery (OARS).7-9 Interestingly, although the prevalence of GERD is increasing, and antireflux surgery has proven to be beneficial for patients, the use of antireflux procedures has declined in the US in recent years.10,11 Although the obesity epidemic and
Gastroesophageal reflux disease (GERD) affects approximately 20% of the population in the US, and its prevalence is increasing worldwide.1 Lifestyle modifications and proton pump inhibitor therapy are effective in the majority of patients and remain the mainstay of GERD treatment. However, some patients will need surgical
Disclosure Information: Nothing to disclose. Received November 7, 2016; Revised December 8, 2016; Accepted December 8, 2016. From the Department of Surgery (Schlottmann, Strassle, Patti), the Center for Esophageal Diseases and Swallowing (Schlottmann, Patti), and the Department of Epidemiology, Gillings School of Global Public Health (Strassle), University of North Carolina at Chapel Hill, Chapel Hill, NC. Correspondence address: Francisco Schlottmann, MD, University of North Carolina at Chapel Hill, 4030 Burnett Womack Building, 101 Manning Drive, CB 7081, Chapel Hill, NC 27599-7081. email: fschlottmann@ hotmail.com
ª 2016 by the American College of Surgeons. Published by Elsevier Inc. All rights reserved.
1
http://dx.doi.org/10.1016/j.jamcollsurg.2016.12.010 ISSN 1072-7515/16
2
Schlottmann et al
Laparoscopic vs Open Fundoplication
Abbreviations and Acronyms
GERD LARS NIS OARS OR
¼ ¼ ¼ ¼ ¼
gastroesophageal reflux disease laparoscopic antireflux surgery National Inpatient Sample open antireflux surgery odds ratio
the rapid increase in the use of bariatric procedures may have contributed to this decline, patients’ and gastroenterologists’ concerns about surgical outcomes may also be related to the decrease in the number of antireflux operations. We hypothesized that in the last decade, LARS was not universally embraced in the US, and that many operations were still performed through a laparotomy, with inferior perioperative outcomes. Therefore, the aims of this study were to assess the national use of LARS in the US, and to compare the perioperative outcomes between laparoscopic and open antireflux procedures in a national cohort.
METHODS A cohort of patients was identified using the National Inpatient Sample (NIS) database between January 1, 2000 and December 31, 2013. The NIS is the largest publicly available, all-payer health care database in the US, and it includes more than 7 million hospitalizations from 1,000 hospitals each year, representing a 20% stratified sample of all hospital discharges in the US. Eligible patients were identified using International Classification of Disease, 9th revision, Clinical Modification (ICD-9CM) diagnostic and procedural codes. Adult patients (18 years or older) diagnosed with gastroesophageal reflux disease (530.11, 530.81, and 530.85), who underwent either an open (44.66) or laparoscopic (44.67) fundoplication during their inpatient hospitalization were eligible for inclusion. Comorbidities of interest included hypertension (401 to 401.9 and 402 to 402.91), primary and secondary diabetes (249 to 249.91 and 250 to 250.93), obesity (278 to 278.8), renal insufficiency (585 to 585.9), coronary artery disease (414 to 414.9), peripheral vascular disease (443 to 443.9), COPD (491 to 492.8), and sleep apnea (327.23). Surgical outcomes of interest were inpatient mortality, postoperative complications during index hospitalization, length of stay, and total charges (excluding operating room time costs). Postoperative complications included venous thromboembolism (415.11, 453.40 to 453.42, and V12.51), wound complications (998.13, 998.30 to 998.32, and 998.83), infection (54.91, 86.04, 567.22, 569.5, 995.9 to 995.99, 996.64, 998.5 to 998.59, and 999.3 to 999.39), esophageal perforation (42.82 and
J Am Coll Surg
530.4), bleeding (99.0 to 99.09, 998.11, and 998.12), shock (998.0 to 998.09), cardiac failure (410 to 410.9, 428 to 428.9), renal failure (38.95, 39.95, 584 to 584.9, 586, and V45.11), and respiratory failure (31.1 to 31.29, 96.04, 96.05, 96.7 to 96.72, and 799.1). A composite complication (ie at least 1 postoperative complication) was also analyzed. Statistical analyses Patient demographics, hospital characteristics, and procedure type were compared across surgery type (laparoscopic vs open) using descriptive statistics. Unadjusted, bivariate analyses of mortality, length of stay, hospital charges, and complication incidence across surgical approach were also assessed using chi-square and Wilcoxon-Mann-Whitney tests. The yearly incidence of laparoscopic fundoplication was estimated using Poisson regression and was expressed as the number of laparoscopic procedures per 100 patients undergoing fundoplication. Missing data for sex (n ¼ 228, 0.3%), race/ethnicity (n ¼ 17,714, 23.5%), primary insurance (n ¼ 321, 0.4%), household income (n ¼ 1,696, 2.3%), admission type (n ¼ 17,181, 22.7%), hospital teaching status (n ¼ 330, 0.4%), bed size (n ¼ 330, 0.4%), inpatient mortality (n ¼ 26, 0.0%), length of stay (n ¼ 2, 0.0%), and hospital charges (n ¼ 1,374, 1.8%) were estimated using Markov Chain Monte Carlo multiple imputation (n ¼ 40). A noninformative prior, 200 burn-in iterations and 100 iterations between imputations was specified. The Markov Chain Monte Carlo models included the variables with missing data plus laparoscopic procedure, other postoperative complications, admission year, admission type, age, comorbidities, hospital region, and hospital surgical volume. Hospital surgical volume was categorized as <10 operations a year (low volume), 10 to 25 operations a year (intermediate volume), and >25 operations a year (high volume). Variables estimates were not rounded or bounded. Main effect multivariable analyses on the potential effect on patient outcomes of laparoscopic surgery, compared with open procedures, were performed using linear and logistic regression, where appropriate, on the imputed datasets. Models were adjusted for admit year, age, sex, race/ethnicity, comorbidities, primary insurance, household income, admit type, hospital region, surgical volume, teaching status, and size. Patient age was modeled as linear variable, as determined by functional form assessment, and centered at 50 years old. A value of p < 0.05 was considered significant for all the statistical methods. All analyses were performed using SAS software version 9.4 (SAS Inc).
Vol.
-,
No.
-, -
2017
Schlottmann et al
Results A total of 75,544 patients were included, with 44,089 LARS (58.4%) and 31,455 OARS (41.6%). There were no meaningful differences in age, sex, race, and presence of comorbidities between patients undergoing laparoscopic and open procedures. Urban teaching hospitals were more likely to perform laparoscopic rather than open procedures (54.4% vs 46.0%, p < 0.0001). Patient and hospital characteristics, stratified by surgical approach, are described in Table 1. The number of fundoplication procedures decreased between 2000 and 2013, with 7,059 procedures occurring in 2000 and only 4,514 occurring in 2013 (Fig. 1A). However, the yearly rate of LARS significantly increased, from 24.8 LARS per 100 procedures in 2000 to 84.3 LARS per 100 procedures in 2013, p < 0.0001 (Fig. 1B). Overall, there were 7,616 (10.1%) complications during the index hospitalization. Open procedures, compared with laparoscopic procedures, had a higher incidence of postoperative wound complications (0.6% vs 0.2%, p < 0.0001), infection (1.7% vs 0.8%, p < 0.0001), esophageal perforation (1.3% vs 0.6%, p < 0.0001), bleeding (1.3% vs 0.6%, p < 0.0001), cardiac failure (2.8% vs 2.0%, p < 0.0001), renal failure (2.7% vs 1.5%, p < 0.0001), respiratory failure (2.7% vs 1.5%, p < 0.0001), shock (0.2% vs 0.1%), and inpatient mortality (0.7% vs 0.3%, p < 0.0001). Patients undergoing laparoscopic procedures had a higher incidence of venous thromboembolism (2.3% vs 1.9%, p < 0.0001) compared with those having open procedures. The median length of stay was 3 days (interquartile range 2 to 6 days) for open procedures and 2 days (interquartile range 1 to 3 days) for laparoscopic procedures, p < 0.0001. Incidence of postoperative complications, length of stay, and total hospital charges among adults undergoing fundoplication with laparoscopic and open surgery are summarized in Table 2. After adjusting for patient and hospital characteristics, patients undergoing laparoscopic surgery were less likely to present postoperative venous thromboembolism (odds ratio [OR] 0.82, 95% CI 0.73, 0.92), wound complications (OR 0.30, 95% CI 0.22, 0.39), infection (OR 0.33, 95% CI 0.28, 0.38), esophageal perforation (OR 0.43, 95% CI 0.36, 0.51), bleeding (OR 0.33, 95% CI 0.30, 0.36), cardiac failure (OR 0.67, 95% CI 0.60, 0.75), renal failure (OR 0.48, 95% CI 0.42, 0.56), respiratory failure (OR 0.48, 95% CI 0.43, 0.54), shock (OR 0.41, 95% CI 0.32, 0.53), and inpatient mortality (OR 0.46, 95% CI 0.32, 0.53). On average, the laparoscopic approach reduced length of stay by 2.08 days (95% CI 2.16, 1.99 days), and decreased hospital charges by $9,530 (95% CI $10.32, $8.74) (Table 3).
Laparoscopic vs Open Fundoplication
3
DISCUSSION The aims of this study were to assess the national use of LARS in the US, and to compare the perioperative outcomes between laparoscopic and open procedures. We found that during the study period, 58.4% of the patients undergoing surgery for GERD had a laparoscopic operation, and LARS was associated with less postoperative morbidity and mortality, shorter length of hospital stay, and lower hospital charges. The rate of adoption of LARS in the US has been quite slow. We do know that in 2003 only 25% of all the antireflux operations were performed through a minimally invasive approach. This is quite impressive considering that the first LARS was reported by Bernard Dallemagne and colleagues5 in Belgium in 1991, and that this technique was rapidly adopted in most academic centers in the US. Between 2003 and 2005 there was a very steep increase in the number of LARS, from 25% to 75%. Subsequently, the rate of increase slowed down considerably, and between 2005 and 2013, only 10% more of antireflux operations were performed laparoscopically. As a result, in 2013, 15% of all antireflux operations were still performed through an open approach. This finding may be due to older surgeons who never embraced laparoscopy, or to factors that might have led surgeons to choose an open rather than a laparoscopic approach (complicated patients, redo-surgery, or patients with multiple previous abdominal operations). We do hope that by 2020, this percentage will be even lower because it is clear that laparoscopy has become the standard approach for many gastrointestinal procedures because it has proven to be effective, while causing less patient discomfort and allowing faster convalescence and return to productive activity. In addition, for patients undergoing antireflux operations, LARS has also shown to be related with better perioperative outcomes. Nilsson and colleagues12 conducted a randomized clinical trial of laparoscopic vs open fundoplication and reported that LARS was associated with less need for analgesics, better respiratory function, and shorter hospital stay. Similarly, in another randomized clinical trial, postoperative pain, analgesic requirement, time to solid food intake, hospital stay, and recovery time were reduced after LARS.7 Ha˚kanson and associates13 also showed that complication rates were lower, and hospital length of stay and time off work were shorter after laparoscopic fundoplication. In light of these results, we would have expected laparoscopy to be adopted widely and early for the surgical treatment of GERD in the US. However, as shown by the trend in the use of laparoscopy, 25 years after the first laparoscopic fundoplication, 15% of
4
Schlottmann et al
J Am Coll Surg
Laparoscopic vs Open Fundoplication
Table 1. Distribution of Patient and Hospital Characteristics among Adult Patients Undergoing Fundoplication Between 2000 and 2013 (N ¼ 75,544) Characteristic
Sex, n (%) Male Female Age, y, mean (95% CI) Race/ethnicity, n (%) Non-Hispanic white Non-Hispanic black Hispanic Other Primary insurance, n (%) Private Public Other/self-pay Household income, n (%)* Low Medium High Highest Comorbidities, n (%) Hypertension Diabetes Obesity Renal insufficiency Coronary artery disease Peripheral vascular disease COPD Sleep apnea Elective admission, n (%) Surgical volume, n (%) Low Intermediate High Hospital size, n (%) Small Medium Large Hospital type, n (%) Urban, teaching Urban, nonteaching Rural, nonteaching Hospital region, n (%) Northeast Midwest South West
Laparoscopic, n ¼ 44,089 (58.4%)
Open, n ¼ 31,455 (41.6%)
15,132 (34.5) 28,774 (65.5) 55 (29e80)
11,770 (37.5) 19,640 (62.5) 54 (28e80)
29,667 1,786 2,200 1,410
29,761 1,152 1,246 608
(84.6) (5.1) (6.3) (4.0)
23,601 (53.8) 18,378 (41.9) 1,924 (4.4)
(86.8) (5.1) (5.5) (2.7)
17,664 (56.4) 12,277 (39.2) 1,379 (4.4)
8,652 11,708 11,336 11,408
(20.1) (27.2) (26.3) (26.5)
4,369 8,413 8,649 9,313
(14.2) (27.4) (28.1) (30.3)
16,501 3,802 4,937 560 3,050 368 676 1,651 26,883
(37.4) (8.6) (11.2) (1.3) (6.9) (0.8) (1.5) (3.7) (85.6)
9.923 2,196 2,903 241 1,883 191 533 399 22,052
(31.6) (7.0) (9.2) (0.8) (6.0) (0.6) (1.7) (1.3) (81.8)
2,404 (5.5) 5,522 (12.5) 36,163 (82.0)
1,565 (5.0) 3,268 (10.4) 26,622 (84.6)
5,097 (11.6) 10,530 (24.0) 28,189 (64.3)
3,692 (11.8) 7,685 (24.5) 20,021 (63.8)
23,828 (54.4) 14,951 (34.1) 5,037 (11.5)
14,435 (46.0) 11,910 (37.9) 5,053 (16.1)
6,958 11,255 14,338 11,538
4,497 8,199 11,572 7,187
(15.8) (25.5) (32.5) (26.2)
(14.3) (26.1) (36.8) (22.9)
*Between 2000 and 2002, household income was characterized by the following quartiles: $1 to $24,999 (low), $25,000 to $34,999 (medium), $35,000 to $44,999 (high), and $45,000 and above (highest); from 2003 onward, income was characterized into quartiles within each ZIP code.
Vol.
-,
No.
-, -
2017
Schlottmann et al
Figure 1. (A) Number of fundoplication procedures per year, stratified by laparoscopic and open procedures and (B) yearly rate of laparoscopic procedures per 100 patients (solid line) and 95% confidence interval (dashed line) among patients undergoing fundoplication.
antireflux operations in the US were still performed through an open approach. Interestingly, slow nationwide implementation of laparoscopy in US has also been seen in colorectal surgery. Laparoscopic colectomy, as compared with open colectomy, has been associated with less postoperative morbidity and a shorter hospital stay, while maintaining the oncologic profile in patients with colon cancer.14,15 Open colectomy has also been associated with an increase in health care costs, use of health care services, and longer recovery time, as shown by the number of days off from work.16 Regardless of these data that clearly demonstrate the advantage of the laparoscopic approach for colon surgery, a recent study has shown that the percentage of open colectomies in the US is still 64.3% with NIS data (period 2006 to 2012) and 51% when the American College of Surgeons NSQIP data are analyzed (period 2006 to 2013).17 In our cohort, LARS was associated with a lower incidence of postoperative complications, a lower rate of
Laparoscopic vs Open Fundoplication
5
inpatient mortality (0.3% vs 0.7%), and a shorter length of hospital stay. Concordantly, a recent analysis of the NSQIP data showed excellent perioperative results after laparoscopic fundoplication, with an overall morbidity rate of 3.8% and 30-day mortality rate of 0.19%.18 Remarkably, similar perioperative outcomes have been shown with the NSQIP data for both appendectomy and cholecystectomy, 2 common operations that are widely considered basic laparoscopic procedures. Although considered a more complex operation, the comparison of LARS outcomes with these data shows similar morbidity (4.46% for appendectomy and 3.1% for cholecystectomy) and 30-day mortality (0.07% for appendectomy and 0.27% for cholecystectomy).19,20 Some opponents of LARS argue that it is less effective than the open approach, and the risk of treatment failure outweighs the short-term benefits related to minimally invasive surgery. Nevertheless, previous reports have reported no significant differences in long-term reflux control and patient satisfaction with surgery between laparoscopic and open fundoplication.8,13 In fact, a recent meta-analysis showed better control of long-term reflux symptoms with LARS as compared with open surgery.9 In addition to better postoperative outcomes, we found that LARS was more cost-effective when compared with open fundoplication. After adjustment for patient and hospital characteristics, laparoscopy reduced hospital costs, on average, by $9,530 per patient. Similarly, Varela and coworkers21 showed that mean hospital costs were significantly higher in patients undergoing open fundoplication ($16,700 vs $9,600). Indirect costs, such as impaired ability to work, workplace absence, and intangible costs of pain and suffering, are difficult to measure, but may also favor the laparoscopic fundoplication. For example, while Heikkinen and colleagues22 reported that hospital costs were similar between LARS ($2,981) and OARS ($3,140), they found that total costs were lower with LARS as a result of an earlier return to work ($7,506 vs $13,118). Additionally, the incidence of incisional hernias after a laparotomy is about 12.8% (range 0% to 35.6%),23 increasing patient discomfort and costs for the health care system. Ha˚kanson and associates13 reported that 7.5% of the patients who underwent open fundoplication required reoperations for incisional hernias. Overall, it seems that costs are significantly reduced with LARS. This study has several limitations. The NIS does not link hospital records, so patient outcomes, including complications, readmission, and mortality, occurring after the initial hospital discharge are unable to be measured. In addition, acquisition of accurate total charge data is difficult in surgical patients because operative time is not
6
Schlottmann et al
J Am Coll Surg
Laparoscopic vs Open Fundoplication
Table 2. Incidence of Postoperative Complications, Length of Stay, and Total Hospital Charges among Adults Undergoing Fundoplication Laparoscopic, n ¼ 44,089 (58.4%)
Characteristic
Postoperative complication, n (%) Venous thromboembolism Wound complication Infection Esophageal perforation Bleeding Cardiac failure Renal failure Respiratory failure Shock Mortality Any complication, n (%)z Length of stay, d, median (IQR) Charges, thousands, median (IQR)
1,031 91 364 273 1,011 875 499 657 43 108 3,639 2 25
(2.3) (0.2) (0.8) (0.6) (2.3) (2.0) (1.1) (1.5) (0.1) (0.3) (8.3) (1e3) (17e42)
Open, n ¼ 31,455 (41.6%)
590 175 546 409 1,592 873 473 863 47 221 3,977 3 19
(1.9) (0.6) (1.7) (1.3) (5.1) (2.8) (1.5) (2.7) (0.2) (0.7) (12.6) (2e6) (13e33)
p Value*
<0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y 0.04y <0.0001y <0.0001y <0.0001y <0.0001y
*Calculated using chi-square and Wilcoxon-Mann-Whitney tests, where appropriate. y p < 0.05. z At least 1 postoperative complication (compared with no complications). IQR, interquartile range.
available, and it contributes significantly to total charges. There is also potential for coding errors in a large administrative database. However, the large sample size of this study should minimize any effect of coding errors. Some factors might have led surgeons to choose an open approach (redo-surgery or patients with multiple
previous abdominal operations). Unfortunately, this information cannot be retrieved with the NIS dataset and therefore, we were not able to adjust for it. In these cases, the rate of complications might also be higher due to surgical complexity. Despite these limitations, large databases provide data on thousands of patients and reflect
Table 3. Crude and Adjusted Odds Ratios and Changes in Estimates of Laparoscopic Procedures, Compared with Open, for Postoperative Complications, Length of Stay, and Hospital Charges Characteristic
Postoperative complication Venous thromboembolism Wound complications Infection Esophageal perforation Bleeding Cardiac failure Renal failure Respiratory failure Shock Mortality Any complicationz Change in estimate, (95% CI) Length of stay, d Charges, thousands
Crude Odds ratio (95% CI)
1.25 0.37 0.47 0.47 0.44 0.71 0.75 0.54 0.65 0.35 0.62
(1.13, (0.29, (0.41, (0.41, (0.41, (0.65, (0.66, (0.48, (0.43, (0.28, (0.59,
1.39) 0.48) 0.54) 0.55) 0.48) 0.78) 0.85) 0.59) 0.99) 0.44) 0.65)
4.45 (4.56, 4.34) 7.11 (8.19, 6.03)
p Value
<0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y 0.04y <0.0001y <0.0001y <0.0001y <0.0001y
Adjusted* Odds ratio (95% CI)
0.82 0.30 0.33 0.43 0.33 0.67 0.48 0.48 0.49 0.41 0.46
(0.73, (0.22, (0.28, (0.36, (0.30, (0.60, (0.42, (0.43, (0.31, (0.32, (0.44,
0.92) 0.39) 0.38) 0.51) 0.36) 0.75) 0.56) 0.54) 0.79) 0.53) 0.49)
2.08 (2.16, 1.99) 9.53 (10.32, 8.74)
p Value
0.0009y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y 0.003y <0.0001y <0.0001y <0.0001y <0.0001y
*Models were adjusted for admission year, age, sex, race/ethnicity, insurance type, income, comorbidities, admission type, surgical volume, hospital size, location/teaching status, and region. y p < 0.05. z At least 1 postoperative complication (compared with no complications).
Vol.
-,
No.
-, -
2017
Schlottmann et al
the outcomes of surgical care provided in the community as a whole rather than at specialized centers in the US.
CONCLUSIONS The results of this study show that the use of laparoscopy for the surgical treatment of GERD has significantly increased in the last decade in US, but 15% of all antireflux operations are still performed by a laparotomy. Laparoscopic antireflux surgery, as compared with open surgery, is associated with better perioperative outcomes and lower costs for the health care system. Considering the many studies that have shown that LARS is as effective as OARS, laparoscopic fundoplication should be the standard of care for the surgical treatment of GERD. Author Contributions Study conception and design: Schlottmann, Strassle, Patti Acquisition of data: Schlottmann, Strassle, Patti Analysis and interpretation of data: Schlottmann, Strassle, Patti Drafting of manuscript: Schlottmann, Strassle, Patti Critical revision: Schlottmann, Strassle, Patti REFERENCES 1. El-Serag HN, Sweet S, Winchester CC, et al. Update on the epidemiology of gastro-esophageal reflux disease: a systematic review. Gut 2014;63:871e880. 2. Dallemagne B, Weerts J, Markiewicz S, et al. Clinical results of laparoscopic fundoplication at ten years after surgery. Surg Endosc 2006;20:159e165. 3. Broeders JA, Rijnhart-de Jong HG, Draaisma WA, et al. Ten-year outcome of laparoscopic and conventional Nissen fundoplication: randomized clinical trial. Ann Surg 2009;250:698e706. 4. Morgenthal CB, Shane MD, Stival A, et al. The durability of laparoscopic Nissen fundoplication: 11-year outcomes. J Gastrointest Surg 2007;11:693e700. 5. Dallemagne B, Weerts JM, Jehaes C, et al. Laparoscopic Nissen fundoplication: preliminary report. Surg Laparosc Endosc 1991;1:138e143. 6. Geagea T. Laparoscopic Nissen’s fundoplication: preliminary report on ten cases. Surg Endosc 1991;5:170e173. 7. Ackroyd R, Watson DI, Majeed AW, et al. Randomized clinical trial of laparoscopic versus open fundoplication for gastrooesophageal reflux disease. Br J Surg 2004;91:975e982. 8. Peters MJ, Mukhtar A, Yunus RM, et al. Meta-analysis of randomized clinical trials comparing open and laparoscopic antireflux surgery. Am J Gastroenterol 2009;104:1548e1561. 9. Qu H, Liu Y, He QS. Short- and long-term results of laparoscopic versus open anti-reflux surgery: a systematic review and meta-analysis of randomized controlled trials. J Gastrointest Surg 2014;18:1077e1086.
Laparoscopic vs Open Fundoplication
7
10. Wang YR, Dempsey DT, Richter JE. Trends and perioperative outcomes of inpatient antireflux surgery in the United States, 1993-2006. Dis Esophagus 2011;24:215e223. 11. Khan F, Maradey-Romero C, Ganocy S, et al. Utilisation of surgical fundoplication for patients with gastro-oesophageal reflux disease in the USA has declined rapidly between 2009 and 2013. Aliment Pharmacol Ther 2016;43: 1124e1131. 12. Nilsson G, Larsson S, Johnsson F. Randomized clinical trial of laparoscopic versus open fundoplication: blind evaluation of recovery and discharge period. Br J Surg 2000;87:873e878. 13. Ha˚kanson BS, Thor KB, Thorell A, Ljungqvist O. Open vs laparoscopic partial posterior fundoplication. A prospective randomized trial. Surg Endosc 2007;21:289e298. 14. Wilson MZ, Hollenbeak CS, Stewart DB. Laparoscopic colectomy is associated with a lower incidence of postoperative complications than open colectomy: a propensity scorematched cohort analysis. Colorectal Dis 2014;16:382e389. 15. Stormark K, Søreide K, Søreide JA, et al. Nationwide implementation of laparoscopic surgery for colon cancer: shortterm outcomes and long-term survival in a population-based cohort. Surg Endosc 2016;30:4853e4864. 16. Crawshaw BP, Chien HL, Augestad KM, Delaney CP. Effect of laparoscopic surgery on health care utilization and costs in patients who undergo colectomy. JAMA Surg 2015;150: 410e415. 17. Schlussel AT, Delaney CP, Maykel JA, et al. A national database analysis comparing the Nationwide Inpatient Sample and American College of Surgeons National Surgical Quality Improvement Program in laparoscopic vs open colectomies: inherent variance may impact outcomes. Dis Colon Rectum 2016;59:843e854. 18. Niebisch S, Fleming FJ, Galey KM, et al. Perioperative risk of laparoscopic fundoplication: safer than previously reportedanalysis of the American College of Surgeons National Surgical Quality Improvement Program 2005 to 2009. J Am Coll Surg 2012;215:61e68. 19. Ingraham AM, Cohen ME, Bilimoria KY, et al. Comparison of outcomes after laparoscopic versus open appendectomy for acute appendicitis at 222 ACS NSQIP hospitals. Surgery 2010;148:625e635. 20. Ingraham AM, Cohen ME, Ko CY, Hall BL. A current profile and assessment of North American cholecystectomy: results from the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg 2010;211: 176e186. 21. Varela JE, Hinojosa MW, Nguyen NT. Laparoscopic improves perioperative outcomes of antireflux surgery at US academic centers. Am J Surg 2008;196:989e993. 22. Heikkinen TJ, Haukipuro K, Koivukangas P, et al. Comparison of costs between laparoscopic and open Nissen fundoplication: a prospective randomized study with a 3-month followup. J Am Coll Surg 1999;188:368e376. 23. Bosanquet DC, Ansell J, Abdelrahman T, et al. Systematic review and meta-regression of factors affecting midline incisional hernia rates: analysis of 14,618 patients. Plos One 2015;10: e0138745.
MD,
Paula D Strassle,
MSPH,
Marco G Patti,
MD, FACS
Laparoscopic antireflux surgery (LARS) has proven to be as effective as open antireflux surgery (OARS), but it is associated with a shorter hospital stay and a faster recover. The aims of this study were to assess the national use of LARS in the US and to compare the perioperative outcomes between laparoscopic and open antireflux procedures in a national cohort. STUDY DESIGN: A retrospective population-based analysis was performed using the National Inpatient Sample for the period 2000 to 2013. The study included adult patients (18 years and older) diagnosed with gastroesophageal reflux disease (GERD), who underwent either laparoscopic or open fundoplication. Multivariable linear and logistic regression, adjusted for patient demographics, comorbidities, and hospital characteristics were used to assess the effect of the laparoscopic approach on patient outcomes. RESULTS: A total of 75,544 patients were included, with 44,089 having LARS (58.4%) and 31,455 having OARS (41.6%). The rate of laparoscopic procedures increased from 24.8 LARS per 100 procedures in 2000, to 84.3 LARS per 100 procedures in 2013 (p < 0.0001). Patients undergoing laparoscopic surgery were less likely to experience postoperative venous thromboembolism, wound complications, infection, esophageal perforation, bleeding, cardiac failure, renal failure, respiratory failure, shock, and inpatient mortality. On average, the laparoscopic approach reduced length of stay by 2.1 days, and decreased hospital charges by $9,530. CONCLUSIONS: The use of the laparoscopic approach for the surgical treatment of GERD has increased significantly in the last decade in the US. This approach is associated with lower morbidity and mortality, shorter hospital stay, and lower costs for the health care system. (J Am Coll Surg 2017;-:1e7. 2016 by the American College of Surgeons. Published by Elsevier Inc. All rights reserved.)
BACKGROUND:
intervention because they have only partial control of symptoms, do not want to be on long-term medical treatment, or suffer complications related to proton pump inhibitor therapy. The most commonly performed antireflux operation is the Nissen fundoplication (360 degrees), which has long-term success in about 80% to 90% of patients.2-4 Laparoscopic antireflux surgery (LARS) was first reported in 19915,6 and has since become widely accepted because of the clear advantages of a minimally invasive approach (decreased pain, shorter hospital stay, faster recovery, and almost complete absence of incisional hernia), and because it is as effective as open antireflux surgery (OARS).7-9 Interestingly, although the prevalence of GERD is increasing, and antireflux surgery has proven to be beneficial for patients, the use of antireflux procedures has declined in the US in recent years.10,11 Although the obesity epidemic and
Gastroesophageal reflux disease (GERD) affects approximately 20% of the population in the US, and its prevalence is increasing worldwide.1 Lifestyle modifications and proton pump inhibitor therapy are effective in the majority of patients and remain the mainstay of GERD treatment. However, some patients will need surgical
Disclosure Information: Nothing to disclose. Received November 7, 2016; Revised December 8, 2016; Accepted December 8, 2016. From the Department of Surgery (Schlottmann, Strassle, Patti), the Center for Esophageal Diseases and Swallowing (Schlottmann, Patti), and the Department of Epidemiology, Gillings School of Global Public Health (Strassle), University of North Carolina at Chapel Hill, Chapel Hill, NC. Correspondence address: Francisco Schlottmann, MD, University of North Carolina at Chapel Hill, 4030 Burnett Womack Building, 101 Manning Drive, CB 7081, Chapel Hill, NC 27599-7081. email: fschlottmann@ hotmail.com
ª 2016 by the American College of Surgeons. Published by Elsevier Inc. All rights reserved.
1
http://dx.doi.org/10.1016/j.jamcollsurg.2016.12.010 ISSN 1072-7515/16
2
Schlottmann et al
Laparoscopic vs Open Fundoplication
Abbreviations and Acronyms
GERD LARS NIS OARS OR
¼ ¼ ¼ ¼ ¼
gastroesophageal reflux disease laparoscopic antireflux surgery National Inpatient Sample open antireflux surgery odds ratio
the rapid increase in the use of bariatric procedures may have contributed to this decline, patients’ and gastroenterologists’ concerns about surgical outcomes may also be related to the decrease in the number of antireflux operations. We hypothesized that in the last decade, LARS was not universally embraced in the US, and that many operations were still performed through a laparotomy, with inferior perioperative outcomes. Therefore, the aims of this study were to assess the national use of LARS in the US, and to compare the perioperative outcomes between laparoscopic and open antireflux procedures in a national cohort.
METHODS A cohort of patients was identified using the National Inpatient Sample (NIS) database between January 1, 2000 and December 31, 2013. The NIS is the largest publicly available, all-payer health care database in the US, and it includes more than 7 million hospitalizations from 1,000 hospitals each year, representing a 20% stratified sample of all hospital discharges in the US. Eligible patients were identified using International Classification of Disease, 9th revision, Clinical Modification (ICD-9CM) diagnostic and procedural codes. Adult patients (18 years or older) diagnosed with gastroesophageal reflux disease (530.11, 530.81, and 530.85), who underwent either an open (44.66) or laparoscopic (44.67) fundoplication during their inpatient hospitalization were eligible for inclusion. Comorbidities of interest included hypertension (401 to 401.9 and 402 to 402.91), primary and secondary diabetes (249 to 249.91 and 250 to 250.93), obesity (278 to 278.8), renal insufficiency (585 to 585.9), coronary artery disease (414 to 414.9), peripheral vascular disease (443 to 443.9), COPD (491 to 492.8), and sleep apnea (327.23). Surgical outcomes of interest were inpatient mortality, postoperative complications during index hospitalization, length of stay, and total charges (excluding operating room time costs). Postoperative complications included venous thromboembolism (415.11, 453.40 to 453.42, and V12.51), wound complications (998.13, 998.30 to 998.32, and 998.83), infection (54.91, 86.04, 567.22, 569.5, 995.9 to 995.99, 996.64, 998.5 to 998.59, and 999.3 to 999.39), esophageal perforation (42.82 and
J Am Coll Surg
530.4), bleeding (99.0 to 99.09, 998.11, and 998.12), shock (998.0 to 998.09), cardiac failure (410 to 410.9, 428 to 428.9), renal failure (38.95, 39.95, 584 to 584.9, 586, and V45.11), and respiratory failure (31.1 to 31.29, 96.04, 96.05, 96.7 to 96.72, and 799.1). A composite complication (ie at least 1 postoperative complication) was also analyzed. Statistical analyses Patient demographics, hospital characteristics, and procedure type were compared across surgery type (laparoscopic vs open) using descriptive statistics. Unadjusted, bivariate analyses of mortality, length of stay, hospital charges, and complication incidence across surgical approach were also assessed using chi-square and Wilcoxon-Mann-Whitney tests. The yearly incidence of laparoscopic fundoplication was estimated using Poisson regression and was expressed as the number of laparoscopic procedures per 100 patients undergoing fundoplication. Missing data for sex (n ¼ 228, 0.3%), race/ethnicity (n ¼ 17,714, 23.5%), primary insurance (n ¼ 321, 0.4%), household income (n ¼ 1,696, 2.3%), admission type (n ¼ 17,181, 22.7%), hospital teaching status (n ¼ 330, 0.4%), bed size (n ¼ 330, 0.4%), inpatient mortality (n ¼ 26, 0.0%), length of stay (n ¼ 2, 0.0%), and hospital charges (n ¼ 1,374, 1.8%) were estimated using Markov Chain Monte Carlo multiple imputation (n ¼ 40). A noninformative prior, 200 burn-in iterations and 100 iterations between imputations was specified. The Markov Chain Monte Carlo models included the variables with missing data plus laparoscopic procedure, other postoperative complications, admission year, admission type, age, comorbidities, hospital region, and hospital surgical volume. Hospital surgical volume was categorized as <10 operations a year (low volume), 10 to 25 operations a year (intermediate volume), and >25 operations a year (high volume). Variables estimates were not rounded or bounded. Main effect multivariable analyses on the potential effect on patient outcomes of laparoscopic surgery, compared with open procedures, were performed using linear and logistic regression, where appropriate, on the imputed datasets. Models were adjusted for admit year, age, sex, race/ethnicity, comorbidities, primary insurance, household income, admit type, hospital region, surgical volume, teaching status, and size. Patient age was modeled as linear variable, as determined by functional form assessment, and centered at 50 years old. A value of p < 0.05 was considered significant for all the statistical methods. All analyses were performed using SAS software version 9.4 (SAS Inc).
Vol.
-,
No.
-, -
2017
Schlottmann et al
Results A total of 75,544 patients were included, with 44,089 LARS (58.4%) and 31,455 OARS (41.6%). There were no meaningful differences in age, sex, race, and presence of comorbidities between patients undergoing laparoscopic and open procedures. Urban teaching hospitals were more likely to perform laparoscopic rather than open procedures (54.4% vs 46.0%, p < 0.0001). Patient and hospital characteristics, stratified by surgical approach, are described in Table 1. The number of fundoplication procedures decreased between 2000 and 2013, with 7,059 procedures occurring in 2000 and only 4,514 occurring in 2013 (Fig. 1A). However, the yearly rate of LARS significantly increased, from 24.8 LARS per 100 procedures in 2000 to 84.3 LARS per 100 procedures in 2013, p < 0.0001 (Fig. 1B). Overall, there were 7,616 (10.1%) complications during the index hospitalization. Open procedures, compared with laparoscopic procedures, had a higher incidence of postoperative wound complications (0.6% vs 0.2%, p < 0.0001), infection (1.7% vs 0.8%, p < 0.0001), esophageal perforation (1.3% vs 0.6%, p < 0.0001), bleeding (1.3% vs 0.6%, p < 0.0001), cardiac failure (2.8% vs 2.0%, p < 0.0001), renal failure (2.7% vs 1.5%, p < 0.0001), respiratory failure (2.7% vs 1.5%, p < 0.0001), shock (0.2% vs 0.1%), and inpatient mortality (0.7% vs 0.3%, p < 0.0001). Patients undergoing laparoscopic procedures had a higher incidence of venous thromboembolism (2.3% vs 1.9%, p < 0.0001) compared with those having open procedures. The median length of stay was 3 days (interquartile range 2 to 6 days) for open procedures and 2 days (interquartile range 1 to 3 days) for laparoscopic procedures, p < 0.0001. Incidence of postoperative complications, length of stay, and total hospital charges among adults undergoing fundoplication with laparoscopic and open surgery are summarized in Table 2. After adjusting for patient and hospital characteristics, patients undergoing laparoscopic surgery were less likely to present postoperative venous thromboembolism (odds ratio [OR] 0.82, 95% CI 0.73, 0.92), wound complications (OR 0.30, 95% CI 0.22, 0.39), infection (OR 0.33, 95% CI 0.28, 0.38), esophageal perforation (OR 0.43, 95% CI 0.36, 0.51), bleeding (OR 0.33, 95% CI 0.30, 0.36), cardiac failure (OR 0.67, 95% CI 0.60, 0.75), renal failure (OR 0.48, 95% CI 0.42, 0.56), respiratory failure (OR 0.48, 95% CI 0.43, 0.54), shock (OR 0.41, 95% CI 0.32, 0.53), and inpatient mortality (OR 0.46, 95% CI 0.32, 0.53). On average, the laparoscopic approach reduced length of stay by 2.08 days (95% CI 2.16, 1.99 days), and decreased hospital charges by $9,530 (95% CI $10.32, $8.74) (Table 3).
Laparoscopic vs Open Fundoplication
3
DISCUSSION The aims of this study were to assess the national use of LARS in the US, and to compare the perioperative outcomes between laparoscopic and open procedures. We found that during the study period, 58.4% of the patients undergoing surgery for GERD had a laparoscopic operation, and LARS was associated with less postoperative morbidity and mortality, shorter length of hospital stay, and lower hospital charges. The rate of adoption of LARS in the US has been quite slow. We do know that in 2003 only 25% of all the antireflux operations were performed through a minimally invasive approach. This is quite impressive considering that the first LARS was reported by Bernard Dallemagne and colleagues5 in Belgium in 1991, and that this technique was rapidly adopted in most academic centers in the US. Between 2003 and 2005 there was a very steep increase in the number of LARS, from 25% to 75%. Subsequently, the rate of increase slowed down considerably, and between 2005 and 2013, only 10% more of antireflux operations were performed laparoscopically. As a result, in 2013, 15% of all antireflux operations were still performed through an open approach. This finding may be due to older surgeons who never embraced laparoscopy, or to factors that might have led surgeons to choose an open rather than a laparoscopic approach (complicated patients, redo-surgery, or patients with multiple previous abdominal operations). We do hope that by 2020, this percentage will be even lower because it is clear that laparoscopy has become the standard approach for many gastrointestinal procedures because it has proven to be effective, while causing less patient discomfort and allowing faster convalescence and return to productive activity. In addition, for patients undergoing antireflux operations, LARS has also shown to be related with better perioperative outcomes. Nilsson and colleagues12 conducted a randomized clinical trial of laparoscopic vs open fundoplication and reported that LARS was associated with less need for analgesics, better respiratory function, and shorter hospital stay. Similarly, in another randomized clinical trial, postoperative pain, analgesic requirement, time to solid food intake, hospital stay, and recovery time were reduced after LARS.7 Ha˚kanson and associates13 also showed that complication rates were lower, and hospital length of stay and time off work were shorter after laparoscopic fundoplication. In light of these results, we would have expected laparoscopy to be adopted widely and early for the surgical treatment of GERD in the US. However, as shown by the trend in the use of laparoscopy, 25 years after the first laparoscopic fundoplication, 15% of
4
Schlottmann et al
J Am Coll Surg
Laparoscopic vs Open Fundoplication
Table 1. Distribution of Patient and Hospital Characteristics among Adult Patients Undergoing Fundoplication Between 2000 and 2013 (N ¼ 75,544) Characteristic
Sex, n (%) Male Female Age, y, mean (95% CI) Race/ethnicity, n (%) Non-Hispanic white Non-Hispanic black Hispanic Other Primary insurance, n (%) Private Public Other/self-pay Household income, n (%)* Low Medium High Highest Comorbidities, n (%) Hypertension Diabetes Obesity Renal insufficiency Coronary artery disease Peripheral vascular disease COPD Sleep apnea Elective admission, n (%) Surgical volume, n (%) Low Intermediate High Hospital size, n (%) Small Medium Large Hospital type, n (%) Urban, teaching Urban, nonteaching Rural, nonteaching Hospital region, n (%) Northeast Midwest South West
Laparoscopic, n ¼ 44,089 (58.4%)
Open, n ¼ 31,455 (41.6%)
15,132 (34.5) 28,774 (65.5) 55 (29e80)
11,770 (37.5) 19,640 (62.5) 54 (28e80)
29,667 1,786 2,200 1,410
29,761 1,152 1,246 608
(84.6) (5.1) (6.3) (4.0)
23,601 (53.8) 18,378 (41.9) 1,924 (4.4)
(86.8) (5.1) (5.5) (2.7)
17,664 (56.4) 12,277 (39.2) 1,379 (4.4)
8,652 11,708 11,336 11,408
(20.1) (27.2) (26.3) (26.5)
4,369 8,413 8,649 9,313
(14.2) (27.4) (28.1) (30.3)
16,501 3,802 4,937 560 3,050 368 676 1,651 26,883
(37.4) (8.6) (11.2) (1.3) (6.9) (0.8) (1.5) (3.7) (85.6)
9.923 2,196 2,903 241 1,883 191 533 399 22,052
(31.6) (7.0) (9.2) (0.8) (6.0) (0.6) (1.7) (1.3) (81.8)
2,404 (5.5) 5,522 (12.5) 36,163 (82.0)
1,565 (5.0) 3,268 (10.4) 26,622 (84.6)
5,097 (11.6) 10,530 (24.0) 28,189 (64.3)
3,692 (11.8) 7,685 (24.5) 20,021 (63.8)
23,828 (54.4) 14,951 (34.1) 5,037 (11.5)
14,435 (46.0) 11,910 (37.9) 5,053 (16.1)
6,958 11,255 14,338 11,538
4,497 8,199 11,572 7,187
(15.8) (25.5) (32.5) (26.2)
(14.3) (26.1) (36.8) (22.9)
*Between 2000 and 2002, household income was characterized by the following quartiles: $1 to $24,999 (low), $25,000 to $34,999 (medium), $35,000 to $44,999 (high), and $45,000 and above (highest); from 2003 onward, income was characterized into quartiles within each ZIP code.
Vol.
-,
No.
-, -
2017
Schlottmann et al
Figure 1. (A) Number of fundoplication procedures per year, stratified by laparoscopic and open procedures and (B) yearly rate of laparoscopic procedures per 100 patients (solid line) and 95% confidence interval (dashed line) among patients undergoing fundoplication.
antireflux operations in the US were still performed through an open approach. Interestingly, slow nationwide implementation of laparoscopy in US has also been seen in colorectal surgery. Laparoscopic colectomy, as compared with open colectomy, has been associated with less postoperative morbidity and a shorter hospital stay, while maintaining the oncologic profile in patients with colon cancer.14,15 Open colectomy has also been associated with an increase in health care costs, use of health care services, and longer recovery time, as shown by the number of days off from work.16 Regardless of these data that clearly demonstrate the advantage of the laparoscopic approach for colon surgery, a recent study has shown that the percentage of open colectomies in the US is still 64.3% with NIS data (period 2006 to 2012) and 51% when the American College of Surgeons NSQIP data are analyzed (period 2006 to 2013).17 In our cohort, LARS was associated with a lower incidence of postoperative complications, a lower rate of
Laparoscopic vs Open Fundoplication
5
inpatient mortality (0.3% vs 0.7%), and a shorter length of hospital stay. Concordantly, a recent analysis of the NSQIP data showed excellent perioperative results after laparoscopic fundoplication, with an overall morbidity rate of 3.8% and 30-day mortality rate of 0.19%.18 Remarkably, similar perioperative outcomes have been shown with the NSQIP data for both appendectomy and cholecystectomy, 2 common operations that are widely considered basic laparoscopic procedures. Although considered a more complex operation, the comparison of LARS outcomes with these data shows similar morbidity (4.46% for appendectomy and 3.1% for cholecystectomy) and 30-day mortality (0.07% for appendectomy and 0.27% for cholecystectomy).19,20 Some opponents of LARS argue that it is less effective than the open approach, and the risk of treatment failure outweighs the short-term benefits related to minimally invasive surgery. Nevertheless, previous reports have reported no significant differences in long-term reflux control and patient satisfaction with surgery between laparoscopic and open fundoplication.8,13 In fact, a recent meta-analysis showed better control of long-term reflux symptoms with LARS as compared with open surgery.9 In addition to better postoperative outcomes, we found that LARS was more cost-effective when compared with open fundoplication. After adjustment for patient and hospital characteristics, laparoscopy reduced hospital costs, on average, by $9,530 per patient. Similarly, Varela and coworkers21 showed that mean hospital costs were significantly higher in patients undergoing open fundoplication ($16,700 vs $9,600). Indirect costs, such as impaired ability to work, workplace absence, and intangible costs of pain and suffering, are difficult to measure, but may also favor the laparoscopic fundoplication. For example, while Heikkinen and colleagues22 reported that hospital costs were similar between LARS ($2,981) and OARS ($3,140), they found that total costs were lower with LARS as a result of an earlier return to work ($7,506 vs $13,118). Additionally, the incidence of incisional hernias after a laparotomy is about 12.8% (range 0% to 35.6%),23 increasing patient discomfort and costs for the health care system. Ha˚kanson and associates13 reported that 7.5% of the patients who underwent open fundoplication required reoperations for incisional hernias. Overall, it seems that costs are significantly reduced with LARS. This study has several limitations. The NIS does not link hospital records, so patient outcomes, including complications, readmission, and mortality, occurring after the initial hospital discharge are unable to be measured. In addition, acquisition of accurate total charge data is difficult in surgical patients because operative time is not
6
Schlottmann et al
J Am Coll Surg
Laparoscopic vs Open Fundoplication
Table 2. Incidence of Postoperative Complications, Length of Stay, and Total Hospital Charges among Adults Undergoing Fundoplication Laparoscopic, n ¼ 44,089 (58.4%)
Characteristic
Postoperative complication, n (%) Venous thromboembolism Wound complication Infection Esophageal perforation Bleeding Cardiac failure Renal failure Respiratory failure Shock Mortality Any complication, n (%)z Length of stay, d, median (IQR) Charges, thousands, median (IQR)
1,031 91 364 273 1,011 875 499 657 43 108 3,639 2 25
(2.3) (0.2) (0.8) (0.6) (2.3) (2.0) (1.1) (1.5) (0.1) (0.3) (8.3) (1e3) (17e42)
Open, n ¼ 31,455 (41.6%)
590 175 546 409 1,592 873 473 863 47 221 3,977 3 19
(1.9) (0.6) (1.7) (1.3) (5.1) (2.8) (1.5) (2.7) (0.2) (0.7) (12.6) (2e6) (13e33)
p Value*
<0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y 0.04y <0.0001y <0.0001y <0.0001y <0.0001y
*Calculated using chi-square and Wilcoxon-Mann-Whitney tests, where appropriate. y p < 0.05. z At least 1 postoperative complication (compared with no complications). IQR, interquartile range.
available, and it contributes significantly to total charges. There is also potential for coding errors in a large administrative database. However, the large sample size of this study should minimize any effect of coding errors. Some factors might have led surgeons to choose an open approach (redo-surgery or patients with multiple
previous abdominal operations). Unfortunately, this information cannot be retrieved with the NIS dataset and therefore, we were not able to adjust for it. In these cases, the rate of complications might also be higher due to surgical complexity. Despite these limitations, large databases provide data on thousands of patients and reflect
Table 3. Crude and Adjusted Odds Ratios and Changes in Estimates of Laparoscopic Procedures, Compared with Open, for Postoperative Complications, Length of Stay, and Hospital Charges Characteristic
Postoperative complication Venous thromboembolism Wound complications Infection Esophageal perforation Bleeding Cardiac failure Renal failure Respiratory failure Shock Mortality Any complicationz Change in estimate, (95% CI) Length of stay, d Charges, thousands
Crude Odds ratio (95% CI)
1.25 0.37 0.47 0.47 0.44 0.71 0.75 0.54 0.65 0.35 0.62
(1.13, (0.29, (0.41, (0.41, (0.41, (0.65, (0.66, (0.48, (0.43, (0.28, (0.59,
1.39) 0.48) 0.54) 0.55) 0.48) 0.78) 0.85) 0.59) 0.99) 0.44) 0.65)
4.45 (4.56, 4.34) 7.11 (8.19, 6.03)
p Value
<0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y 0.04y <0.0001y <0.0001y <0.0001y <0.0001y
Adjusted* Odds ratio (95% CI)
0.82 0.30 0.33 0.43 0.33 0.67 0.48 0.48 0.49 0.41 0.46
(0.73, (0.22, (0.28, (0.36, (0.30, (0.60, (0.42, (0.43, (0.31, (0.32, (0.44,
0.92) 0.39) 0.38) 0.51) 0.36) 0.75) 0.56) 0.54) 0.79) 0.53) 0.49)
2.08 (2.16, 1.99) 9.53 (10.32, 8.74)
p Value
0.0009y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y <0.0001y 0.003y <0.0001y <0.0001y <0.0001y <0.0001y
*Models were adjusted for admission year, age, sex, race/ethnicity, insurance type, income, comorbidities, admission type, surgical volume, hospital size, location/teaching status, and region. y p < 0.05. z At least 1 postoperative complication (compared with no complications).
Vol.
-,
No.
-, -
2017
Schlottmann et al
the outcomes of surgical care provided in the community as a whole rather than at specialized centers in the US.
CONCLUSIONS The results of this study show that the use of laparoscopy for the surgical treatment of GERD has significantly increased in the last decade in US, but 15% of all antireflux operations are still performed by a laparotomy. Laparoscopic antireflux surgery, as compared with open surgery, is associated with better perioperative outcomes and lower costs for the health care system. Considering the many studies that have shown that LARS is as effective as OARS, laparoscopic fundoplication should be the standard of care for the surgical treatment of GERD. Author Contributions Study conception and design: Schlottmann, Strassle, Patti Acquisition of data: Schlottmann, Strassle, Patti Analysis and interpretation of data: Schlottmann, Strassle, Patti Drafting of manuscript: Schlottmann, Strassle, Patti Critical revision: Schlottmann, Strassle, Patti REFERENCES 1. El-Serag HN, Sweet S, Winchester CC, et al. Update on the epidemiology of gastro-esophageal reflux disease: a systematic review. Gut 2014;63:871e880. 2. Dallemagne B, Weerts J, Markiewicz S, et al. Clinical results of laparoscopic fundoplication at ten years after surgery. Surg Endosc 2006;20:159e165. 3. Broeders JA, Rijnhart-de Jong HG, Draaisma WA, et al. Ten-year outcome of laparoscopic and conventional Nissen fundoplication: randomized clinical trial. Ann Surg 2009;250:698e706. 4. Morgenthal CB, Shane MD, Stival A, et al. The durability of laparoscopic Nissen fundoplication: 11-year outcomes. J Gastrointest Surg 2007;11:693e700. 5. Dallemagne B, Weerts JM, Jehaes C, et al. Laparoscopic Nissen fundoplication: preliminary report. Surg Laparosc Endosc 1991;1:138e143. 6. Geagea T. Laparoscopic Nissen’s fundoplication: preliminary report on ten cases. Surg Endosc 1991;5:170e173. 7. Ackroyd R, Watson DI, Majeed AW, et al. Randomized clinical trial of laparoscopic versus open fundoplication for gastrooesophageal reflux disease. Br J Surg 2004;91:975e982. 8. Peters MJ, Mukhtar A, Yunus RM, et al. Meta-analysis of randomized clinical trials comparing open and laparoscopic antireflux surgery. Am J Gastroenterol 2009;104:1548e1561. 9. Qu H, Liu Y, He QS. Short- and long-term results of laparoscopic versus open anti-reflux surgery: a systematic review and meta-analysis of randomized controlled trials. J Gastrointest Surg 2014;18:1077e1086.
Laparoscopic vs Open Fundoplication
7
10. Wang YR, Dempsey DT, Richter JE. Trends and perioperative outcomes of inpatient antireflux surgery in the United States, 1993-2006. Dis Esophagus 2011;24:215e223. 11. Khan F, Maradey-Romero C, Ganocy S, et al. Utilisation of surgical fundoplication for patients with gastro-oesophageal reflux disease in the USA has declined rapidly between 2009 and 2013. Aliment Pharmacol Ther 2016;43: 1124e1131. 12. Nilsson G, Larsson S, Johnsson F. Randomized clinical trial of laparoscopic versus open fundoplication: blind evaluation of recovery and discharge period. Br J Surg 2000;87:873e878. 13. Ha˚kanson BS, Thor KB, Thorell A, Ljungqvist O. Open vs laparoscopic partial posterior fundoplication. A prospective randomized trial. Surg Endosc 2007;21:289e298. 14. Wilson MZ, Hollenbeak CS, Stewart DB. Laparoscopic colectomy is associated with a lower incidence of postoperative complications than open colectomy: a propensity scorematched cohort analysis. Colorectal Dis 2014;16:382e389. 15. Stormark K, Søreide K, Søreide JA, et al. Nationwide implementation of laparoscopic surgery for colon cancer: shortterm outcomes and long-term survival in a population-based cohort. Surg Endosc 2016;30:4853e4864. 16. Crawshaw BP, Chien HL, Augestad KM, Delaney CP. Effect of laparoscopic surgery on health care utilization and costs in patients who undergo colectomy. JAMA Surg 2015;150: 410e415. 17. Schlussel AT, Delaney CP, Maykel JA, et al. A national database analysis comparing the Nationwide Inpatient Sample and American College of Surgeons National Surgical Quality Improvement Program in laparoscopic vs open colectomies: inherent variance may impact outcomes. Dis Colon Rectum 2016;59:843e854. 18. Niebisch S, Fleming FJ, Galey KM, et al. Perioperative risk of laparoscopic fundoplication: safer than previously reportedanalysis of the American College of Surgeons National Surgical Quality Improvement Program 2005 to 2009. J Am Coll Surg 2012;215:61e68. 19. Ingraham AM, Cohen ME, Bilimoria KY, et al. Comparison of outcomes after laparoscopic versus open appendectomy for acute appendicitis at 222 ACS NSQIP hospitals. Surgery 2010;148:625e635. 20. Ingraham AM, Cohen ME, Ko CY, Hall BL. A current profile and assessment of North American cholecystectomy: results from the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg 2010;211: 176e186. 21. Varela JE, Hinojosa MW, Nguyen NT. Laparoscopic improves perioperative outcomes of antireflux surgery at US academic centers. Am J Surg 2008;196:989e993. 22. Heikkinen TJ, Haukipuro K, Koivukangas P, et al. Comparison of costs between laparoscopic and open Nissen fundoplication: a prospective randomized study with a 3-month followup. J Am Coll Surg 1999;188:368e376. 23. Bosanquet DC, Ansell J, Abdelrahman T, et al. Systematic review and meta-regression of factors affecting midline incisional hernia rates: analysis of 14,618 patients. Plos One 2015;10: e0138745.