Abstract
Introduction:
Paraesophageal hernias readily affect the elderly with a median age of presentation between 65 and 75 years. Laparoscopic paraesophageal hernia repair (PEHR) is a technically challenging operation with potential for dire complications. Advanced age and medical comorbidities may heighten perioperative risk and limit surgical candidacy, potentially refusing patients an opportunity toward symptom resolution. Given the increased prevalence in the elderly and associated surgical risks, we aim to assess age as an independent risk factor for perioperative morbidity and mortality after PEHR.
Methods:
A retrospective analysis using a prospectively maintained database assessed patients undergoing PEHR from 2007 to 2018. Patients were stratified by age: Group A (age <65 years), Group B (65≤ age <80 years), and Group C (age ≥80 years). Patient demographics, preoperative symptoms, postoperative outcomes, and mortality rate were analyzed. Barium esophagram was performed on symptomatic postsurgical patients. Recurrence was confirmed radiologically.
Results:
In total, 143 patients underwent laparoscopic (94.4%) or robotic-assisted (5.6%) PEHR. Average age per group was Group A (n = 49) 55.4 years (standard deviation [SD] ±8.91), Group B (n = 76) 71.4 years (SD ±4.40), and Group C (n = 17) 84.1 (years) (SD ±3.37). Group C had significantly higher rates of nonelective surgery (P = .018), preoperative weight loss (P = .014), hypertension (P = .031), ischemic heart disease (P = .001), and cancer (P = .039); preoperative body mass index was significantly lower (P = .048). Charlson comorbidity index differences between groups were significant (2.00 versus 3.61 versus 5.28, P < .001). Median follow-up was 426 days (6–3199). Symptom improvement was seen in 78.3% of patients. Recurrence and reoperation rates were not significantly different between groups. No differences were seen in mortality, length of stay, or postoperative complications between groups.
Conclusions:
PEHR in elderly patients proved to be safe and effective. Avoidance of emergent intervention may be achieved through a judicious elective approach to this anatomic problem. Symptom resolution and quality-of-life improvement can be safely achieved with surgical repair in this patient population, demonstrating that age is truly just a number for PEHR.
Introduction
Paraesophageal hernias (PEHs) are most prevalent in the elderly population and care is often debated regarding indications, method, and timing of surgical correction. PEH development is attributed to progressive stretching and weakening of the phrenoesophageal ligament, accompanied by gradual enlargement of the diaphragmatic hiatus. Patient presentation is variable, ranging from asymptomatic radiographic findings to extremis. The most common symptoms associated with PEH, such as early satiety, regurgitation, vomiting, dysphagia, heartburn, cough, or dyspnea, are often medically controlled or tolerated for years before surgical assessment. Progression invariably leads to a larger hernia over time with much of the stomach or other associated organs located in the chest. 1 Furthermore, symptoms develop in ∼14% of previously asymptomatic patients annually. 2
Indications for repair of a PEH can vary among surgeons and are often individualized between patients. The historical recommendation was for repair of all PEHs, irrespective of symptoms. 3 Over time, some providers became reluctant to refer or proceed with surgical repair due to the perception that the procedure carried a prohibitive risk of perioperative morbidity and mortality. Approach then shifted toward expectant management in the elderly population due to their increased perioperative risk and comorbidity burden. 4 However, emergent repair of PEH, often secondary to gastric volvulus or strangulation, has been associated with up to a sevenfold increase in mortality rate when compared with elective repairs. 5
In recent years, the minimally invasive approach to paraesophageal hernia repair (PEHR) has resulted in a reduction of morbidity. 3 Furthermore, elective laparoscopic PEHRs have been shown to have excellent perioperative outcomes. 3 Recent data also indicated many patients with PEH have symptoms affecting their daily quality of life. 2 Quality-of-life metrics have reported significant improvement after elective laparoscopic PEHR with the majority of patients demonstrating relief of heartburn, dysphagia, regurgitation, dyspnea, and anemia. 4 In the geriatric population in particular, targeted elective repair may be advantageous to circumvent acute presentation requiring emergent intervention.
Published series of PEHR show that the procedure is most frequently performed in patient populations with a median age between 64 and 70 years. 6 The proportion of elderly patients is also rapidly expanding. Data from the U.S. Census Bureau estimated that adults >65 years composed 15.2% of the population in 2016; in 2060, this age group is projected to account for nearly 25% of the population. 7 As PEH disproportionately affects the geriatric population, one can project that the proportion of PEHs requiring surgical management will continue to expand.
The paradigm has recently shifted yet again, with many surgeons now weighing the patient's current symptoms, clinical status, risk of continued symptom development, and acute presentation against potential complications associated with surgical repair. The objective of this study was, therefore, to assess age as an independent risk factor for perioperative morbidity and mortality after laparoscopic PEHR.
Materials and Methods
A retrospective review of a prospectively maintained IRB-approved database was performed to identify patients who underwent laparoscopic or robotic PEHR between April 2007 and July 2018. PEH was defined as a type II, III, or IV hiatal hernia. Patients undergoing revisional PEHR for recurrence were excluded from the primary analysis, but data are provided within a subgrouping for comparison. PEH was diagnosed through barium swallow and upper endoscopy. Collected patient characteristics include age, gender, body mass index (BMI), American Society of Anesthesiologists (ASA) score, medical comorbidities, and elective versus nonelective repair. Nonelective repair was defined as an urgent intervention with surgical repair occurring during that acute admission. Collected symptoms include epigastric or chest pain, dysphagia, heart burn or reflux, weight loss, cough, belching, shortness of breath, and early satiety. Patients were stratified by age: Group A (age <65 years), Group B (65 ≤ age <80 years), and Group C (age ≥80 years). We grouped patients in this manner to differentiate outcomes between early geriatrics and late geriatrics. Several studies have also demonstrated worse perioperative and postoperative outcomes with patients over the age of 80 years when compared with younger patients, and these studies grouped patients in a similar manner.8–12 Intraoperative data collected include hiatal hernia type, operative approach, and fundoplication type. Selection of fundoplication type was based on preoperative manometry. When manometry was unavailable, a partial posterior fundoplication was performed.
Postoperative outcomes included length of stay (LOS), 30-day morbidity, 30-day mortality, recurrence, and need for revisional operation secondary to recurrence. Postoperative complications were ranked using the Clavien–Dindo (CD) classification system. Barium esophagram was performed at the 1-year postoperative visit, and in symptomatic postoperative patients. Radiologic recurrence was defined by anatomic positioning of the gastroesophageal junction above the hiatus on imaging. Symptomatic recurrence was identified by review of clinical notes from postoperative follow-up visits. Symptoms indicative of recurrence included heartburn, reflux, nausea and/or vomiting, and regurgitation. Univariate analysis was performed using Pearson's chi-square test for categorical variables and one-way analysis of variance for scale variables. All P values were considered statistically significant when the associated probability was <0.05. IBM SPSS Statistics version 25 for Windows (IBM, Inc., Armonk, NY) was used for statistical analysis.
Results
A total of 143 identified patients underwent laparoscopic or robotic PEHR. In total, 71.3% of patients were female, average BMI was 27.9 kg/m2. Group A (n = 49) had an average age of 55.4 years (standard deviation [SD] ±8.91), Group B (n = 76) had an average age of 71.4 years (SD ±4.40), and Group C (n = 17) had an average age of 84.1 years (SD ±3.37). Patient demographic data are given in Table 1. Symptomatic indications for surgery included anemia, epigastric or chest pain, dysphagia, heartburn, weight loss, cough, and early satiety. Regarding the anemic indication for surgery, 48 patients in total presented with a complaint of anemia, 5 were found to have an associated gastric ulcer, and only 1 required urgent surgical intervention. Presenting symptoms were not significant between groups, although heartburn, epigastric/chest pain, and regurgitation were the most common. Charlson comorbidity index (CCI) and comorbidities are listed in Table 2. Type III hiatal hernia was the most common type, comprising 93.0% (Table 1). The majority of patients (94.4%) underwent laparoscopic repair, with the remainder (5.6%) undergoing a robotic-assisted approach. Nissen fundoplication was performed in 71.3%, and 28.0% underwent a 270° Toupet fundoplication. One patient (0.7%) underwent a concurrent PEH and sleeve gastrectomy with no fundoplication.
Patient Characteristics
ASA, American Society of Anesthesiologists; BMI, body mass index; DOS, date of surgery.
Medical Comorbidities
CCI, Charlson comorbidity index; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; DM, diabetes mellitus.
Baseline characteristics were compared and demonstrated some variability among studied groups. Rate of nonelective surgery was significantly higher in Group C (2.0% versus 2.6% versus 16.7%, P = .018) (Table 1). There was no significant difference in ASA classification (P = .198), hernia type (P = .320), or type of wrap (P = .089) (Table 1). In the age group >80 years, weight loss as a presenting symptom was significantly higher (10.2% versus 2.6% versus 22.2%, P = .014) and preoperative BMI was significantly lower (28.4 kg/m2 versus 27.0 kg/m2 versus 25.1 kg/m2, P = .048) than other groups (Table 1). In assessing comorbidities, hypertension (P = .031), ischemic heart disease (P = .001), and cancer (P = .039) were higher in Group C (Table 2). Significant differences were demonstrated between groups with respect to CCI (2.00 versus 3.61 versus 5.28, P < .001) (Table 2). The prevalence of liver disease was elevated in Group A (P = .001) (Table 2).
Of the 6 patients who required nonelective PEHR, indications included incarceration (1 in Group A, 1 in Group B, 2 in Group C) and acute blood loss (1 in Group B, 1 in Group C). No 30-day morbidity or mortality occurred within the emergent case group, nor were any symptomatic recurrences noted. One mortality occurred in Group B. This patient suffered cardiac arrest on postoperative day 5, thought to be secondary to a large left-sided hydropneumothorax causing mediastinal shift. There were no statistically significant differences in mortality rate (0.0% versus 1.3% versus 0.0%, P = .642), LOS (2.0 versus 1.8 versus 2.8 days, P = .222), or rate of major postoperative complication (4.0% versus 10.5% versus 5.6%, P = .392) between groups (Table 3). A significant difference in minor complication rates between Groups A, B, and C was observed (4.1% versus 14.5% versus 33.3%, P = .007). Postoperative outcomes are listed in Table 3. CD complications ≥3 are listed in Table 4.
Patient Outcomes
CD, Clavien-Dindo Score; LOS, length of stay.
Clavien–Dindo Complications >3
Resolution of specific symptoms postoperatively is given in Table 5, with 84.4% reporting resolution of heartburn, 90.5% with resolution of cough, and 87.9% reporting improvement in epigastric and chest pain. In total, 66.2% of patients had complete symptom resolution postoperatively, with no significant difference between groups.
Symptom Resolution Postprocedure
Preoperative BMI demonstrated statistical significance between groups (28.4 kg/m2 versus 27.0 kg/m2 versus 25.1 kg/m2, respectively), with Group C having the lowest BMI at the time of surgical intervention (P = .048). Postoperatively, all groups were noted to have small decreases in BMI compared with preoperative measures (−0.41 versus −0.91 versus −0.25), with the smallest decreases seen in Group C, although statistical significance was not achieved (P = .613) (Table 1).
Recurrence rates across groups, based on both symptomatic presentation and evidence of recurrent PEH on upper GI barium swallow, were not statistically significant (26.5% versus 25.3% versus 11.1%, P = .389) (Table 3). Of the 61 patients who completed their 1-year postoperative esophagram, recurrence was detected radiologically in 47 patients. Only 34 of these were clinically relevant. Overall rate of reoperation after initial operation at our institution was 7.7%. Between groups, reoperation for clinically recurrent PEH was 14.2% versus 5.3% versus 0.0% and was not statistically significant (P = .077) (Table 3). Of the 11 patients requiring reoperation, 7 occurred in Group A, 4 in Group B, and none in Group C. Group A patients had a mean time of symptom recurrence 20.6 months postprocedure (range 3–57 months), and had an average time period of 25.3 months between initial and redo PEHR (range 11–59 months). Group B patients had symptomatic recurrence at an average of 10.75 months (range 3–24 months), with a mean time period of 22.5 months (range 3–43 months) between initial and redo PEHR. Table 6 tallies the symptoms complained of during recurrence. There was one 30-day morbidity in Group A, none in Group B. Symptom resolution after redo PEHR in Groups A and B was 71.4% and 100%, respectively. Median duration of follow-up was 321 days.
Postoperative Symptomatic Recurrence
Discussion
In this single-center study, we demonstrated comparable LOS and major 30-day morbidity and mortality rates when comparing octogenarians with younger age groups. Recurrence rates were similar across different age groups. We noted an increased comorbidity burden and CCI score in the octogenarian group, but without increased risk for major postoperative complications. Elderly patients underwent laparoscopic or robotic PEHR with low morbidity and mortality rates that were comparable with younger patient groups. Current surgical guidelines suggest that age should not preclude repair in symptomatic patients. 16 Our findings support this stance. Symptomatic hernias can be surgically corrected regardless of age, addressing both quality of life and reductions in need for emergent surgery.
Spaniolas et al. reviewed a series of 2,681 patients, 11.7% of whom were >80 years. Not surprisingly, the majority of these patients were classified ASA 3 or 4 in comparison with lower ASA scores in the younger cohort. 6 Mortality and serious morbidity were not significantly different in the >80 years group versus the <80 years group, but minor and overall morbidity showed significant differences. 6 Similarly, our series reflected a similar proportion of octagenarians, with 11.9% of patients age 80 years and over. There were no significant differences in ASA between our three patient groups. There were no significant differences in mortality and serious morbidity between each group. However, as in the previous report, the rate of minor complications in the elderly was higher than other groups.
Perioperative risk stratification tools, applied toward surgical selection and management of PEH in the elderly population, have been recommended in the literature. Ballian et al. advocated for routine cardiac testing on the majority of patients before elective and urgent repair when feasible. 4 Any patient >65 years presenting as a potential surgical candidate for PEHR within our institution must undergo cardiac clearance before intervention. In patients with concomitant pulmonary issues, pulmonary consultation and clearance are obtained before PEHR. Proactive optimization of appropriately selected surgical candidates has likely contributed to the low rates of morbidity and mortality within our cohort, specifically in the octogenarian population. This thought is further supported by the comparable morbidity and mortality rates despite the elderly population's higher CCI score and comorbidity burden.
Conservative or expectant management of symptomatic PEH has shown that the risk of requiring emergent or urgent surgical intervention increases over time. 2 Furthermore, repairs performed in a nonelective manner have been shown to have increased mortality rates, ranging from 5.4% to 17%. 2 Poulose et al. specifically looked at PEHR in octogenarians using the National Inpatient Sample, and found that 43% underwent nonelective repair. 17 The overall crude mortality rate was 8.4% for a sample of 1,005 patients >80 years. Similar to other studies, nonelective repair demonstrated a higher mortality rate of 15.7%, as compared with elective repair (2.4%). 17 Our series demonstrated a 1.3% mortality rate with a single death occurring in Group B in the context of elective surgery. Nonelective PEHR occurred in 4.2% of our patients with half of these (2.1%) in the ≥80 years group. We demonstrated no mortalities in this nonelective group, which is lower than many reports in the literature. However, given the increased incidence of risk described with nonelective repair with increasing age, a reasonable approach to decrease mortality associated with urgent repair is to address PEH in an elective manner. Both the literature and our study outcomes have shown that elderly patients do not experience increased postoperative mortality or major morbidity with elective laparoscopic repair.1,4,18
The durability of laparoscopic PEHR has contributed to the ongoing debate regarding the best approach, timing, and utility of PEHR. Recurrence rates after laparoscopic PEHR are broad, with a bulk of reported rates in the literature ranging from 20.5% to 30%. 15 We show a recurrence rate of 23.7% with no differences between age groups, and a majority of patients adequately managed with medical therapy. A multi-institutional randomized study by Oelschlager et al. reported that the anatomic recurrence rate at a median of 5 years was ∼50%. 19 With these elevated rates of recurrence, the definition of a “good outcome” post-PEHR has remained elusive, further contributing to the ongoing debate regarding optimal management of PEH.
Reoperative rates, however, remain low. Our reoperative rate of 7.8% is comparable with that reported by Lazar et al. and Müller-Stich et al.: 9.9% and 8.0%, respectively.15,20 Although a limited number of revisional PEHRs were performed within the scope of this study, we demonstrated symptom resolution at 71.4% in Group A and 100% in Group B. Surgical treatment of recurrent PEH can pose a significant technical challenge. 21 Although analysis of redo PEHR outcomes is beyond the scope of this study, patient selection and surgical experience should be considered when considering revisional PEH surgery.21,22 With this, previous groups have shown comparable outcomes between initial and redo PEHRs, concordant with our results. 22
With respect to surgical approach for primary PEHR, technical approach is not ubiquitous with some groups preferring hernia reduction and some form of gastropexy in both acute and elective settings. 23 We support full hernia reduction with fundoplication. 24 Finally, the use of mesh cruroplasty is of debate regardless of age. Biosynthetic mesh, with reported benefit of reduced short-term PEH recurrence, is our preferred approach.19,25,26
Quality of life is an important consideration in this patient population. In assessing patient-reported outcomes after laparoscopic PEHR, Lazar et al. found that >90% of patients were satisfied with their surgery to the point of recommending it to others. 15 Merzlikin et al. identified patients >70 years of age who underwent PEHR with at least 5 years follow-up postprocedure. Patient quality of life at long-term follow-up was sustained, and found to be significantly improved compared with baseline. 27 Although our study did not assess level of patient satisfaction, we observed excellent symptom resolution with respect to reflux-associated symptoms, such as heartburn, epigastric and chest pain, and regurgitation. In addition, a prior study conducted by our group demonstrated 50% resolution of anemia after PEHR in patients with identifiable Cameron's lesions, as well as statistically significant resolution of anemia in 72.2% of patients without endoscopically identifiable lesions. 28 The symptomatic benefits favor elective repair of PEH.
The BMI–surgical risk curve is U shaped, and recent studies have shown that there is a close relationship between low BMI and postoperative complications.29,30 We noted a statistically significant difference in BMI between Group C and other groups. Despite the lower observed BMI in Group C, there were no increases in morbidity and mortality within this subgroup. This result may indicate that weight loss and failure to thrive could be indolent underlying “symptoms” in this age group of patients with PEH. 31 Furthermore, although statistical significance was not achieved in looking at changes in BMI postoperatively, Group C did have the smallest decrease in BMI postprocedure. Anecdotally, we have noted that patients with symptomatic PEH often suffer from unintentional weight loss secondary to gastrointestinal symptoms. Recent studies looking at frailty in the elderly population have listed unintentional weight loss as a contributing factor to increasingly frailty scores. 31 Studies moving forward could potentially examine the effect of PEHR on weight regain in the elderly frail population.
Limitations to our study include the retrospective nature of the data review, limited sample size, and absence of quality of life and frailty assessments. Poor 1-year follow-up limits the use of routine postoperative testing and extraction of long-term symptomatology assessment. In terms of nonelective repairs, only 6 patients presented in an urgent manner; other studies have a greater proportion of nonelective cases, which may account for our improved morbidity and mortality rates. Regarding dysphagia assessment, we did not use a scored questionnaire to assess preoperative severity and postoperative symptomatic improvement. Use of standardized questionnaires for symptom resolution, quality of life, and frailty would permit objective assessments of improvement postprocedure. Despite these limitations, our study indicates that a minimally invasive approach for PEHR in elderly patients may be safely performed in the setting of increased comorbidity burden, with proper patient selection.
Although age alone should not preclude operative intervention, careful consideration should still be applied while selecting elderly surgical patients. PEHR has proven to be safe and effective in the elderly population, but still has potential for complication and mishap. Addo et al. demonstrated in their study of PEHR in 492 elderly patients a higher rate of intraoperative complications and reoperation rates without differences in mortality or major morbidity. 32 Complications included intraoperative visceral injury or esophageal injury during bougie placement. Although in our own study we demonstrated no difference in reoperation rates between age groups, this potentially affected the relatively fewer patients in Group C. Patient selection should be individualized based on factors such as comorbidity burden, symptom severity, hernia size, risk of progression toward hernia incarceration or strangulation, and revisional versus virgin repair. Perioperative screening tools and optimization are also paramount for elective repairs.
Conclusion
Elective PEHR in the symptomatic elderly population is both safe and effective. Avoidance of emergent intervention may be achieved through a judicious elective approach to this anatomic problem. As our aging population grows, so too will the prevalence of PEH. Surgical planning should weigh the variables of symptoms, outcomes, risk, and quality of life. Symptom and quality-of-life improvements can be achieved with surgical intervention. Age is truly just a number when it comes to PEHR.
Footnotes
Disclosure Statement
Dr. S.H. is a consultant for Stryker Corporation, Intuitive Surgical, Fortimedix Surgical, and Medtronic. Dr. B.J.S. is a consultant for Intuitive Surgical and Boston Scientific. Dr. G.R.J. receives a teaching honorarium from Gore Medical. Drs. J.N.C., K.N., A.M.L., R.D.-P., R.C.B., and T.M. have no conflicts of interest or financial ties to disclose.
Funding Information
No funding was received for this project.
