Impact of Operative Times and Mesh Utilization on Paraesophageal Hernia Repair: Analysis of 30-Day Outcomes from the American College of Surgeons National Surgical Quality Improvement Project Database

Abstract

Background:

Using mesh to buttress the crural repair following a paraesophageal hernia repair remains controversial. This article evaluates recent trends in laparoscopic paraesophageal hernia repairs and analyzes the impact of mesh and operative time on postoperative morbidity.

Methods:

The 2013–2015 American College of Surgeons National Surgical Quality Improvement Project (ACS-NSQIP) database was queried for elective laparoscopic paraesophageal hernia repair with and without mesh. Operative times were grouped into quartiles and statistical analysis was performed using analysis of variance univariate with post hoc testing and multivariate regression modeling. The outcomes of interest were composite morbidity scores and readmission rates within 30 days of surgery.

Results:

The database identified a cohort of 6234 laparoscopic paraesophageal hernia repairs. Mesh was utilized in 42% of cases per year and did not change over the study period (P = .367). Mesh was used 37%, 40%, 43%, and 49% of the time within operative quartiles 1, 2, 3, and 4, respectively (P < .001). Postoperative morbidity and readmission rates for each operative time quartile were 2.8%, 4.1%, 5.42%, and 6.13% (P < .001) and 4.4%, 5%, 6.2%, and 7.6% (P = .001), respectively. Post hoc testing indicated statistically significant differences in postoperative morbidity and readmission rates between quartiles 1 and 3/4. Multivariate regression analysis documented operative time as a risk factor for postoperative morbidities and readmission. Simply using mesh was not directly associated with postoperative morbidity.

Conclusion:

Mesh utilization does not impact postoperative outcomes; however, as operative time increases, the incidence of postoperative morbidity also increases.

Introduction

Due to the size of the hiatal defect, foreshortening of the esophagus and tension along the cruroplasty, paraesophageal hernia repairs are technically challenging.¹ Multiple techniques have been employed since Soresi's original description of a paraesophageal hernia repair in 1919, and include open and minimally invasive approaches through the thorax and abdomen.^2–4 Nevertheless, the key components of any repair remain constant and include adequate mobilization of the esophagus into the abdominal cavity, excision of the hernia sac, tension-free closure of the crural defect, and the addition of a fundoplication.⁵

Despite multiple repair options, paraesophageal recurrence rates remain high following a primary crural closure. These rates range from 15% for open approaches and up to 42% for laparoscopic approaches.⁶ Recurrent hernias typically occur posteriorly along the crural repair secondary to excessive tension or inadequate esophageal length.^7–9 Synthetic and biologic meshes have been used over the last two decades to reinforce the crural repair. Mesh has been proposed as an adjunct to large hiatal defects, while others utilize mesh routinely.¹⁰ However, the efficacy of mesh versus no mesh has not been definitively documented and mesh is associated with potential complications such as mesh erosions, mesh migration, fibrosis, and esophageal stenosis.^11,12 While the use of a crural buttress with biological mesh may decrease short-term recurrence rates, the long-term radiologic recurrence rate is equivalent to a primary repair.^1,13

Currently, laparoscopic repair has become the standard approach for transabdominal paraesophageal hernia repair, given its decreased overall morbidity, decreased postoperative pain, and decreased length of stay compared to the open approach.¹⁴ Whether a procedure is performed laparoscopically or by laparotomy, longer operative times have been associated with an increased incidence of urinary tract infections, wound infections, pneumonias, prolonged intubations, deep vein thrombosis, and sepsis^15–18 To date, there are no studies evaluating the impact of overall operative time on outcomes following laparoscopic paraesophageal hernia repair. This study utilizes the American College of Surgeons National Surgical Quality Improvement Project (ACS-NSQIP) database to quantify the correlation between mesh utilization and operative times and postoperative morbidity following a laparoscopic paraesophageal hernia repair.

Methods

The ACS-NSQIP database for 2013–2015 was retrospectively reviewed for patients aged between 40 and 90 years, who underwent a paraesophageal hernia repair with or without mesh utilization. The following CPT codes were used to isolate laparoscopic abdominal paraesophageal hernia repair with (43282) and without (43281) mesh. Cases marked as emergency or cases with another major concurrent procedure were excluded from the analysis. Similarly, patients who had a preoperative hospital stay longer than 1 day or patients with sepsis or other infections present at the time of surgery were excluded. Cases with missing morbidity data were excluded from the analysis. Reported operative times were transformed into quartiles (80–110, 111–142, 143–185, and 186–360 minutes). Cases with operative times <80 or >360 minutes were excluded to decrease the effect of outlying data points. Excluded cases represented ∼20% of the initial dataset (10% on each tail).

This analysis utilized information provided by the NSQIP database. These variables include demographics, comorbidities, serologies, and preoperative, perioperative, and postoperative metrics and outcomes. Data extraction and statistical analysis were performed using IBM SPSS ver. 24 and G*Power v3.0.10. Univariate analysis was used to determine differences in preoperative variables based on mesh utilization and operative time quartiles. Post hoc testing was applied to determine outcomes based on operative time quartiles and Pearson correlation was utilized to determine relationships between mesh utilization and operative time. The primary outcome of interest was the incidence of composite morbidity and readmission within 30 days between patients who underwent repair with and without mesh utilization. Composite morbidity encompassed any occurrence of an ACS-NSQIP-reported postoperative complication, excluding mortality during the 30-day postoperative period. Additional composite morbidities were defined as follows: wound complications included superficial, deep, and organ space surgical site infections, and wound disruptions; pulmonary complications included pneumonia, unplanned reintubation, and ventilator duration >48 hours; and thromboembolic complications included pulmonary embolism and deep venous thrombosis. The secondary outcome of interest was the effect of operative time on postoperative morbidity and readmission rates following laparoscopic paraesophageal hernia repair. Based on the incidence of the primary and secondary outcomes and the sample size, this study was powered to detect a difference of <2%.

Multivariate regression modeling was used to control for differences in the mesh versus no mesh within the different operative time quartiles. Variables used in regression modeling were derived from differences in preoperative factors between cohorts in univariate analysis. Continuous variables with minimal clinical differences or with >20% missing data were excluded from multivariate models. Multivariate regression models also considered the interaction between mesh utilization and operative time. P values <.05 were considered statistically significant and 95% confidence intervals (CIs) were reported for all odds ratios (ORs).

Results

Between 2013 and 2015, there were 6234 paraesophageal hernia repairs reported to NSQIP that met the inclusion and exclusion criteria for this analysis: 3600 (57.7%) without mesh and 2634 (43.3%) with mesh. The rate of mesh utilization for paraesophageal hernia repair was constant over the study period (Table 1). Patients who underwent a mesh repair were older (64.4 versus 63.2, P < .001), had a lower body mass index (30.4 versus 31.2, P < .001), and had more subjective dyspnea (15.1% versus 12.4%, P = .002), but had a lower incidence of significant weight loss within 6 months of surgery (1.44% versus 2.14%, P = 0.044) compared to those who had a repair without mesh (Table 2). Similarly, patients with longer operative times were older, had a higher rate of mesh utilization and had a higher incidence of weight loss before surgery (Table 3). There was no difference in the average body mass index between operative time quartiles.

Table 1.

Paraesophageal Hernia Repair with Mesh Utilization over the Period of Study as Reported to National Surgical Quality Improvement Project

	Year
Paraesophageal hernia repair reported to NSQQIP	2013	2014	2015
Mesh (CPT 43282), n (%)	781 (43.63)	868 (41.87)	985 (41.54)
No Mesh (CPT 43281), n (%)	1009 (56.36)	1205 (58.12)	1386 (58.45)

Table 2.

Analysis of Variance for Patient Demographics in the Elective Paraesophageal Hernia Repair Group Based on Mesh Utilization

Preoperative demographics for operative time quartiles in minutes	No mesh (n = 3600), value (n)	Mesh (n = 2634), value (n)	P
Age	63.25	64.43	<.001
Body mass index (kg/m²)	31.23	30.44	<.001
Diabetic	10.22 (368)	9.57 (252)	.393
Smoking	8.56 (308)	8.31 (219)	.735
Dyspnea	12.39 (446)	15.07 (397)	.002
Chronic obstructive pulmonary disease	5.5 (198)	6.42 (169)	.129
Congestive heart failure	0.25 (9)	0.38 (10)	.359
Hypertension	52.47 (1889)	54.37 (1432)	.139
Steroid usage	3.64 (131)	4.14 (109)	.312
Weight loss	2.14 (77)	1.44 (38)	.044
Preoperative sepsis/SIRS	0 (0)	0 (0)	N/A
Infectious condition PATOS	0 (0)	0 (0)	N/A

Continuous variables are reported as means. Categorical variables are reported in percentages. Variables with a low incidence were selectively excluded.

PATOS, present at time of surgery; SIRS, systemic inflammatory response syndrome.

Table 3.

Analysis of Variance for Patient Demographics in the Elective Paraesophageal Hernia Repair Group Based on Operative Time Quartile

Preoperative demographics for operative time quartiles in minutes	Quartile 1 (80–110), n = 1598, value (n)	Quartile 2 (111–142), n = 1551, value (n)	Quartile 3 (143–185), n = 1550, value (n)	Quartile 4 (186–360), n = 1535, value (n)	P
Age	62.65	63.99	63.91	64.47	<.001
Body mass index (kg/m²)	30.87	30.82	30.73	31.18	.218
Diabetic	9.76 (156)	8.77 (136)	10.06 (156)	11.21 (172)
Smoking	7.7 (123)	8.12 (126)	9.23 (143)	8.79 (135)	.42
Dyspnea	12.83 (205)	12.89 (200)	13.29 (206)	15.11 (232)	.206
Chronic obstructive pulmonary disease	5.82 (93)	5.54 (86)	6 (93)	6.19 (95)	.89
Congestive heart failure	0.44 (7)	0.26 (4)	0.19 (3)	0.33 (5)	.567
Hypertension	52.19 (834)	52.22 (810)	53.61 (831)	55.11 (846)	.308
Steroid usage	3.5 (56)	4.45 (69)	3.48 (54)	3.97 (61)	.45
Weight loss	0.94 (15)	2.64 (41)	2 (31)	1.82 (28)	.005
Preoperative sepsis/SIRS	0 (0)	0 (0)	0 (0)	0 (0)	N/A
Infectious condition PATOS	0 (0)	0 (0)	0 (0)	0 (0)	N/A
Mesh utilization	37.71 (603)	40.23 (624)	43.23 (670)	48.66 (747)	<.001

Continuous variables are reported as means. Categorical variables are reported in percentages. Variables with a low incidence were selectively excluded.

PATOS, present at time of surgery; SIRS, systemic inflammatory response syndrome.

Univariate analysis of 30-day postoperative outcomes based on mesh utilization showed a significant increase in average operative time (159.27 versus 149.72 minutes, P < .001) of a laparoscopic paraesophageal hernia repair and a trend toward a significantly lower rate of thromboembolic complications (0.53% versus 0.94%, P = .065) for paraesophageal hernia repair with mesh versus without mesh (Table 4). There was no statistically significant difference in composite morbidity (P = .403) or postoperative hospital length of stay (P = .739) based on mesh utilization. Pearson correlation statistics indicated that there was a weak positive correlation between mesh utilization and operative time (0.085, P < .001).

Table 4.

Unadjusted 30-Day Postoperative Outcomes Following Elective Paraesophageal Hernia Repair Group Based on Mesh Utilization

Postoperative outcome for mesh utilization	No mesh (n = 3600), value (n)	Mesh (n = 2634), value (n)	P
Operative time (minutes)	149.72	159.27	<.001
Wound complication	1.33 (48)	0.95 (25)	.163
Pulmonary complication	2.28 (82)	2.24 (59)	.92
Thromboembolic complication	0.94 (34)	0.53 (14)	.065
Reoperation	2.56 (92)	2.05 (54)	.192
Readmission	5.45 (196)	6.19 (163)	.214
Composite morbidity	10.64 (383)	10 (263)	.403
Hospital length of stay	2.30	2.33	.739

Continuous variables are reported as means. Categorical variables are reported in percentages. Variables with an incidence of <0.5% were selectively excluded. Composite morbidity represents a grouping of all reported postoperative morbidities available in NSQIP, including readmission and reoperation during the 30-day postoperative period.

Independent of mesh utilization, as operative time quartile increased, there was an associated increase in the incidence of wound complications (P < .001), pulmonary complications (P = .058), thromboembolic complications (P = .001), readmission (P = .001), and postoperative length of stay (P < .001) (Table 5). Post hoc analysis of univariate outcome and operative time quartile showed a large intergroup difference in hospital length of stay, wound complications, pulmonary complications, thromboembolic complications, and readmission between quartiles 1 and 4. There was a similar increase in wound complications and thromboembolic complications between quartiles 2, 3, and 4 and an increase in readmission rate between quartiles 2 and 4 (Table 6).

Table 5.

Unadjusted 30-Day Postoperative Outcomes Following Elective Paraesophageal Hernia Repair Group Based on Operative Time Quartile

Preoperative demographics for operative time quartiles in minutes	Quartile 1 (80–110), n = 1598, value, n	Quartile 2 (111–142), n = 1551, value, n	Quartile 3 (143–185), n = 1550, value, n	Quartile 4 (186–360), n = 1535, value, n	P
Wound complication	0.56 (9)	0.97 (15)	0.97 (15)	2.21 (34)	<.001
Pulmonary complication	1.63 (26)	2.26 (35)	2.13 (33)	3.06 (47)	.058
Thromboembolic complication	0.5 (8)	0.45 (7)	0.58 (9)	1.56 (24)	.001
Reoperation	1.75 (28)	2.51 (39)	2.19 (34)	2.93 (45)	.164
Readmission	4.38 (70)	4.96 (77)	6.19 (96)	7.56 (116)	.001
Composite morbidity	7.134 (114)	9.1 (141)	11.61 (180)	13.75 (211)	<.001
Hospital length of stay	1.78	2.14	2.41	2.94^a	<.001

All subgroup analyses of differences in postoperative hospital length of stay were significant on Tukey post hoc testing.

Table 6.

Tukey Post Hoc Testing for 30-Day Postoperative Outcomes Following Elective Paraesophageal Hernia Repair Group Based on Operative Time Quartile

Post hoc analysis of postoperative outcomes	Quartile comparison		Difference (%)	P
Wound complication	Quartile 1	Quartile 4	−1.65	<.001
	Quartile 2	Quartile 4	−1.25	.007
	Quartile 3	Quartile 4	−1.25	.007
Pulmonary complication	Quartile 1	Quartile 4	−1.44	.035
Thromboembolic complication	Quartile 1	Quartile 4	−1.06	.004
	Quartile 2	Quartile 4	−1.11	.002
	Quartile 3	Quartile 4	−0.98	.010
Readmission	Quartile 1	Quartile 4	−3.20	.001
Readmission	Quartile 2	Quartile 4	−2.60	.011

Differences reported between first and second referenced operative time quartile.

Multivariate regression analysis of composite morbidity identified a negative association between an increasing operative time quartile and composite morbidity (quartile 2 versus quartile 1 OR 1.283, P = .059; quartile 3 versus quartile 1 OR 1.696, P < .001; and quartile 4 versus quartile 1 OR 2.031, P < .001). There was no association between mesh utilization and composite morbidity (P = .119) (Table 7). Age (OR 1.009, P = .022), diabetes (OR 1.301, P = .042), dyspnea (OR 1.581, P < .001), and weight loss (OR 1.64, P = .058) had a varying degree of negative impact on postoperative morbidity. There was no effect of body mass index or the presence of renal failure on postoperative morbidity.

Table 7.

Multivariate Regression Analysis of 6233 Cases Evaluating the Association of Operative Time Quartiles on Separate 30-Day Postoperative Outcomes

Multivariate regression analysis for composite morbidity	Odds ratio	95% Confidence interval	P
Age (continuous, years)	1.009	1.001–1.017	.022
Body mass index (continuous, kg/m²)	0.999	0.985–1.013	.908
Diabetes	1.301	1.010–1.677	.042
Dyspnea	1.581	1.276–1.958	<.001
Renal failure	4.979	0.33–82.615	.263
Weight loss	1.64	0.983–2.737	.058
Mesh utilization	0.875	0.739–1.035	.119
Operative time versus quartile 1
Quartile 2	1.283	0.991–1.662	.059
Quartile 3	1.696	1.325–2.170	<.001
Quartile 4	2.031	1.595–2.586	<.001

All-cause morbidity term represents a composite score of wound, pulmonary, thromboembolic complications, and reoperation. Readmission and postoperative length of stay are not included in the all-cause morbidity metric.

Discussion

The results of this analysis suggest that individuals undergoing a laparoscopic paraesophageal hernia repair have similar 30-day postoperative outcomes with or without mesh utilization, but are at greater risk for postoperative morbidity as operative time increases. While similar short-term outcomes based on mesh utilization for laparoscopic paraesophageal hernia repairs have been previously proven using the NSQIP database,^19,20 this analysis provides a novel, quantifiable relationship between mesh utilization, operative time, and postoperative morbidity following a laparoscopic paraesophageal hernia repair. Gaining a deeper understanding of the relationship between operative time and postoperative morbidity will allow surgeons to better counsel their patients on postoperative expectations. It will also facilitate postoperative protocols aimed at detecting postoperative morbidity at earlier and potentially less threatening stages.

To date, little has been published quantifying estimated operative times for laparoscopic paraesophageal hernia repairs. The limited data suggest that operative times vary from <170 to >260 minutes based on surgeon experience and hernia complexity, not taking into account mesh utilization.^15,21,22 Comparatively, this analysis suggests that operative times have improved in recent years, averaging 155 minutes, and that mesh utilization increased operative times by ∼10 minutes without a statistically significant impact on 30-day postoperative outcomes based on mesh utilization. The results of this analysis also agree with prior reports correlating increased operative time quartiles with increased postoperative morbidity.^15–18 Most notably, there was a 72% increase in readmission rates and a 92% increase in composite morbidity, comparing the first and fourth operative time quartile on univariate analysis subsequently confirmed on multivariate regression (OR 2.031, CI 1.595–2.586, P < .001). Hospital length of stay was also over 1-day longer comparing the first and fourth quartiles.

The authors believe these increases in morbidity are important to understand when encouraging patients to follow advanced recovery programs aimed at earlier discharge dates and would caution surgeons to treat cases with longer operative times more cautiously than cases with shorter operative times. Expectedly, the interquartile difference in postoperative morbidity is smaller between quartile two and four and quartile three and four, but remains statistically significant. Also, there was no significant difference between quartiles 1 and 2, 1 and 3, or 2 and 3 on univariate analysis, which is likely due to the incremental increase in postoperative morbidity per unit time.

Using operative time as a metric for estimating morbidity within a large dataset provides the added benefit of decreasing variability between hidden patient and surgeon factors. While it is likely that surgeon skill and patient anatomy directly impact operative times, abstracting these values with a large, diverse surgeon pool allows for an objective analysis of the stress on the body of complex laparoscopic surgery. A surgeon can subsequently evaluate their own operative times to determine a relative risk profile for their particular patients. Furthermore, while large databases are limited in available preoperative information such as past surgical history, patients who undergo redo operations will be at a higher risk for adhesions, a longer dissection, and longer operative times. When coupled with the negative association of increased age, presence of diabetes and dyspnea, and preoperative weight loss, the overall frailty of a patient and their susceptibility to postoperative morbidity can be more accurately predicted as operative time increases.^23,24

The limitations of this study stem from the structure of the reporting periods of the ACS-NSQIP dataset. While the ACS-NSQIP dataset provides thousands of cases and hundreds of data points per case, detailed operative information such as location and type of mesh used and whether the referenced case was a paraesophageal hernia recurrence is hidden. Furthermore, follow-up data are limited to 30-day postoperative and likely would not catch complications of mesh placement such as long-term mesh erosion or hernia recurrence. These limitations may confound the conclusion that there are no differences in paraesophageal hernias repaired laparoscopically with or without mesh utilization.

Given the association between longer operative times and an increased incidence of postoperative morbidity and readmission, future studies should focus on techniques to reduce total operative time without compromising the durability of the repair. The authors recognize that total operative time and mesh utilization are both patient and surgeon dependent and that these results should be interpreted relative to a surgeon's existing performance metrics. However, the authors do suggest that if mesh is used, the surgeon should acknowledge that operative times will be incrementally longer, and that the incidence of postoperative morbidity may be proportionally higher versus a repair without mesh.

Conclusion

This analysis suggests that while mesh utilization by itself does not impact postoperative outcomes, as operative time increases, the incidence of postoperative morbidity and rates of readmission also increase.

Footnotes

Acknowledgment

This study was performed without grant support or support from other financial relationships.

Authors' Contributions

Conception: M.S., C.S., and F.B.; data acquisition: M.S. and R.A.; analysis: M.S., I.N.H., and R.A.; article preparation: M.S., I.N.H., C.S., R.A., and F.B.; review: M.S., I.N.H., C.S., R.A., and F.B.

Disclosure Statement

No competing financial interests exist.

References

Oelschlager

, Pellegrini

, Hunter

, et al. Biologic prosthesis to prevent recurrence after laparoscopic paraesophageal hernia repair: Long-term follow-up from a multicenter, prospective, randomized trial. J Am Coll Surg, 2011; 213:461–468.

Soresi

. Diaphragmatic hernia: Its unsuspected frequency: Diagnosis and technique for radical cure. Ann Surg, 1919; 69:254–270.

Cuschieri

, Shimi

, Nathanson

. Laparoscopic reduction, crural repair, and fundoplication of large hiatal hernia. Am J Surg, 1992; 163:425–430.

Congreve

. Laparoscopic paraesophageal hernia repair. J Laparoendosc Surg, 1992; 2:45–48.

Morrow

, Oelschlager

. Laparoscopic paraesophageal hernia repair. Surg Laparosc Endosc Percutan Tech, 2013; 23:446–448.

Hashemi

, Peters

, DeMeester

, et al. Laparoscopic repair of large type III hiatal hernia: Objective followup reveals high recurrence rate. J Am Coll Surg, 2000; 190:553–560; discussion 560–561.

Smith

, Isaacson

, Draganic

, Baladas

, Falk

. Symptomatic and radiological follow-up after para-esophageal hernia repair. Dis Esophagus, 2004; 17:279–284.

Furnée

EJB

, Draaisma

, Broeders

IAMJ

, Gooszen

. Surgical reintervention after failed antireflux surgery: A systematic review of the literature. J Gastrointest Surg, 2009; 13:1539–1549.

Awais

, Luketich

, Schuchert

, et al. Reoperative antireflux surgery for failed fundoplication: An analysis of outcomes in 275 patients. Ann Thorac Surg, 2011; 92:1083–1089; discussion 1089–1090.

10.

Pfluke

, Parker

, Bowers

, Asbun

, Daniel Smith

. Use of mesh for hiatal hernia repair: A survey of SAGES members. Surg Endosc, 2012; 26:1843–1848.

11.

Tatum

, Shalhub

, Oelschlager

, Pellegrini

. Complications of PTFE mesh at the diaphragmatic hiatus. J Gastrointest Surg, 2008; 12:953–957.

12.

Stadlhuber

, Sherif

, Mittal

, et al. Mesh complications after prosthetic reinforcement of hiatal closure: A 28-case series. Surg Endosc, 2009; 23:1219–1226.

13.

Oelschlager

, Pellegrini

, Hunter

, et al. Biologic prosthesis reduces recurrence after laparoscopic paraesophageal hernia repair: A multicenter, prospective, randomized trial. Ann Surg, 2006; 244:481–490.

14.

Kohn

, Price

, DeMeester

, et al. Guidelines for the management of hiatal hernia. Surg Endosc, 2013; 27:4409–4428.

15.

Schauer

, Ikramuddin

, McLaughlin

, et al. Comparison of laparoscopic versus open repair of paraesophageal hernia. Am J Surg, 1998; 176:659–665.

16.

Daley

, Cecil

, Clarke

, Cofer

, Guillamondegui

. How slow is too slow? Correlation of operative time to complications: An analysis from the Tennessee Surgical Quality Collaborative. J Am Coll Surg, 2015; 220:550–558.

17.

Bailey

, Davenport

, Vargas

, Evers

, McKenzie

. Longer operative time: Deterioration of clinical outcomes of laparoscopic colectomy versus open colectomy. Dis Colon Rectum, 2014; 57:616–622.

18.

Scheer

, Martel

, Moloo

, et al. Laparoscopic colon surgery: Does operative time matter?. Dis Colon Rectum, 2009; 52:1746–1752.

19.

Schlottmann

, Strassle

, Patti

. Laparoscopic paraesophageal hernia repair: Utilization rates of mesh in the USA and short-term outcome analysis. J Gastrointest Surg, 2017; 21:1571–1576.

20.

Kubasiak

, Hood

, Daly

, et al. Improved patient outcomes in paraesophageal hernia repair using a laparoscopic approach: A study of the national surgical quality improvement program data. Am Surg, 2014; 80:884–889.

21.

Karmali

, McFadden

, Mitchell

, et al. Primary laparoscopic and open repair of paraesophageal hernias: A comparison of short-term outcomes. Dis Esophagus, 2008; 21:63–68.

22.

Parker

, Rambhajan

, Horsley

, Johanson

, Gabrielsen

, Petrick

. Laparoscopic paraesophageal hernia repair is safe in elderly patients. Surg Endosc, 2017; 31:1186–1191.

23.

Chimukangara

, Frelich

, Bosler

, Rein

, Szabo

, Gould

. The impact of frailty on outcomes of paraesophageal hernia repair. J Surg Res, 2016; 202:259–266.

24.

Spaniolas

, Laycock

, Adrales

, Trus

. Laparoscopic paraesophageal hernia repair: Advanced age is associated with minor but not major morbidity or mortality. J Am Coll Surg, 2014; 218:1187–1192.