Abstract
Background
Hiatal hernia is a common surgical pathology. Such hernias can be found incidentally and patients may opt for an initial nonoperative approach though many will pursue surgery after symptom progression. Data on the effects of age on the outcomes of hiatal hernia repair may help inform this decision-making process.
Methods
The TriNetX database was queried for all adult patients undergoing hiatal hernia repair from 2000 to 2023. Patients were divided into elective and emergent cohorts on the basis of diagnosis codes indicating obstruction or gangrene. Patients aged 80–89 were compared against those aged 65–79 in unadjusted analysis. Logistic regression models controlling for additional health history covariates were created to calculate odds ratios for primary outcomes.
Results
There were 2310 octogenarians and 15,295 seniors who underwent elective hiatal hernia repair, and 406 octogenarians and 1462 seniors who underwent emergent repair during the study period. The vast majority of patients in both groups underwent minimally invasive operations. In the elective cohort, octogenarians had higher rates of mortality, malnutrition, sepsis, respiratory failure, pneumonia, DVT, blood transfusion, and discharge to nursing facility. In the emergent cohort, octogenarians had higher rates of mortality, malnutrition, sepsis, and respiratory failure. The odds ratios for mortality in the elective and emergent cohorts were 3.9 (95% CI 3.1–5.0) and 3.5 (95% CI 2.1–5.6), respectively.
Conclusion
Octogenarians are at a meaningfully increased risk for mortality and morbidity after both elective and emergent hiatal hernia repair compared to senior-aged patients. Greater consideration should be given to surgical repair prior to the 8th decade of life.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00464-024-10956-y.
Keywords: Hiatal, Paraesophageal, Geriatric, Frailty, Asymptomatic
Hiatal hernia is a common surgical pathology, with over 60,000 repairs performed in the United States in 2018 [1]. Many such hernias are found incidentally, leaving the surgeon and patient with a difficult management decision [2]. Given the association between age and size of hiatal hernia, this will become a more frequent dilemma as the population continues to age [3]. Data on the effects of age on the outcomes of hiatal hernia repair may help inform the decision-making process.
The management of hiatal hernia has undergone many iterations over the last century. In 1919, Dr. Soresi published the first account of elective repair of hiatal hernias and advocated for the routine repair of even very small hernias [4]. Fifty years later, some surgeons advocated for observation of minimally symptomatic hernias while others, including Dr. Belsey, advocated for the repair of every paraesophageal hernia [5, 6]. This recommendation was largely born out of concern regarding the risk of gastric volvulus with strangulation.
In 2013, the Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) published its Guidelines for the Management of Hiatal Hernia [7]. One of its recommendations was that “Routine elective repair of completely asymptomatic paraesophageal hernias may not always be indicated.” This recommendation was made on the basis of a Markov Monte Carlo decision analytic model by Stylopoulos et al. that found the average 65 year-old patient with a minimally symptomatic paraesophageal hernia would benefit more from a strategy of watchful waiting than operative management [8].
This model assumed 13.87% of the observation population would crossover into surgical management due to symptom progression each year. However, it did not account for worse outcomes in this scenario due to elective repair at an older age. Additionally, outcomes of paraesophageal hernia repair have improved as cases centralize to high volume centers and care improves [9].
This study was undertaken to delineate the modern outcomes of hiatal hernia repair in an elderly patient cohort compared to a younger cohort to help describe the potential consequences of delaying elective hiatal hernia repair until later in life.
Material and methods
This population-based, retrospective cohort study was conducted using TriNetX (Cambridge, Massachusetts), a federated health network research database. TriNetX is a multicenter electronic health research network that provides de-identified electronic health record (EHR) data from participating health care organizations (HCOs). Clinical variables are derived directly through the EHRs. Within TriNetX, we utilized data extracted from the Research Network, which is composed of 68 HCOs. ICD-9, ICD-10, and CPT codes were used to extrapolate diagnostic and procedural information. Details of codes utilized for patient selection are described in Supplemental Material 1.
All adult patients (age ≥ 18 years) with a diagnosis of hiatal hernia from 2000 to 2023 in the US were included in this study. Two separate cohorts were created: an emergent hiatal hernia repair cohort and an elective cohort. The emergent cohort was created by selecting patients with a diagnosis code indicating obstruction or gangrene and hiatal hernia operation within 3 days of diagnosis. The elective cohort was created by selecting patients with a diagnosis of hiatal hernia without obstruction or gangrene and hiatal hernia operation at any time after diagnosis. The hiatal hernia repair is considered the index event in this analysis. Within each cohort, two separate groups were created based on age at index operation: the senior group, which included patients aged 65–79 years, and the octogenarian group, which included patients aged 80–89 years (Fig. 1).
Fig. 1.
Flowchart for selection of patients who underwent non-emergent or emergent hiatal hernia repair
The primary outcomes of this study were mortality, acute respiratory failure, and sepsis. The secondary outcomes included reoperation, malnutrition, pneumonia, deep vein thrombosis (DVT), pulmonary embolism (PE), blood transfusion, need for physical therapy, and discharge to a nursing facility. All outcomes reported in this manuscript were measured at 90 days.
All statistical analysis was performed in R software version 4.3.1 with data obtained from the TriNetX platform. For all primary and secondary outcomes, the absolute risk difference (ARD) between the octogenarian group and the senior group was calculated for the elective and emergent cohorts separately. Additionally, 95% confidence intervals (CI) for these ARDs were calculated. For each primary outcome in each cohort, a logistic regression model was generated, modeling the log odds of the outcome of interest as a function of the following covariates: octogenarian status, history of obesity, history of ischemic heart disease, history of chronic lower respiratory disease, and history of tobacco use. The additional health history covariates were chosen to control for their effects. Using these logistic regression models, the odds ratio (OR) for octogenarians as compared to seniors was estimated for each primary outcome, with accompanying Wald Cis and p-values to test the null hypothesis that the odds of each outcome was no different between octogenarians and seniors. To control for multiplicity, we employed a Bonferroni correction across our six comparisons in order to fix our family-wise error rate to be 0.05, and thus we performed each individual test at the 0.00625 level.
This study received a waiver of review from the institutional review board as the TriNetX database includes only aggregated counts and statistical summaries of de-identified information.
Results
A total of 124,453,162 patients were assessed in the Research Network. Of these, 46,924 were adults who underwent elective hiatal hernia repair, and 3462 were adults who underwent emergent hiatal hernia repair. In the non-emergent cohort, 2310 patients (13.2%) were octogenarians and 15,295 patients (86.8%) were seniors; in the emergent cohort, 406 patients (21.7%) were octogenarians and 1462 patients (78.3%) were seniors. Patient characteristics of the two cohorts are presented in Table 1. With the exception of age, the two cohorts were quite similar. Predictably, octogenarians in both the elective and emergent groups more frequently had hypertensive disease, ischemic heart disease, cerebrovascular disease, and failure to thrive. They had lower rates of nicotine or tobacco use than seniors. Within the elective cohort, octogenarians had a lower rate of obesity. TriNetX does not provide BMI data.
Table 1.
Baseline characteristics of octogenarians (80–89) and seniors (65–79) in the non-emergent and emergent hiatal hernia repair cohorts
| Elective HH repair | Emergent HH repair | |||
|---|---|---|---|---|
| n (%, 95% CI) | n (%, 95% CI) | |||
| Octogenarian | Senior | Octogenarian | Senior | |
| n = 2310 | n = 15,295 | n = 406 | n = 1462 | |
| Age, mean | 82.6 | 72.6 | 83 | 71.9 |
| Females | 1539 (66.6%) | 10,692 (69.9%) | 241 (59.4%) | 963 (65.9%) |
| White | 1910 (82.7%) | 12,634 (82.6%) | 326 (80.3%) | 1188 (81.3%) |
| African American | 50 (2.2%) | 569 (3.7%) | 6 (1.5%) | 41 (2.8%) |
| Other | 58 (2.5%) | 407 (2.7%) | 16 (3.9%) | 31 (2.1%) |
| Comorbidities | ||||
| Overweight and obese | 479 (20.7%) | 5048 (33.0%) | 71 (17.5%) | 434 (20.7%) |
| Type 2 diabetes | 417 (18.1%) | 2867 (18.7%) | 69 (17.0%) | 266 (18.2%) |
| Hypertensive disease | 1800 (77.9%) | 10,345 (67.6% | 334 (82.3%) | 1031 (70.5%) |
| Ischemic heart disease | 796 (34.5%) | 3543 (23.1%) | 151 (37.2%) | 351 (24.0%) |
| Cerebrovascular disease | 414 (17.9%) | 1471 (9.6%) | 75 (18.5%) | 144 (9.9%) |
| Chronic lower respiratory disease | 768 (33.2%) | 5014 (32.8%) | 129 (31.8%) | 465 (31.8%) |
| Nicotine or tobacco use | 98 (4.2%) | 1133 (7.4%) | 20 (4.9%) | 123 (8.4%) |
| Underweight | 9 (0.4%) | 37 (0.2%) | 4 (1.0%) | 5 (0.3%) |
| Failure to thrive | 53 (2.3%) | 135 (0.9%) | 18 (4.4%) | 27 (1.8%) |
HH hiatal hernia, CI confidence interval
The majority of operations were completed with a laparoscopic approach in both the elective and emergent cohorts (Table 2).
Table 2.
Cases by operative approach within both the elective and emergent cohorts
| Elective HH | Emergent HH | |||
|---|---|---|---|---|
| n (%) | n (%) | |||
| Octogenarian | Senior | Octogenarian | Senior | |
| Open abdominal approach | 106 (4.58%) | 502 (3.28%) | 51 (12.56%) | 164 (11.21%) |
| Open thoracic approach | 20 (0.86%) | 229 (1.49%) | 10 (2.46%) | 44 (3.00%) |
| Thoracoabdominal approach | 7 (0.30%) | 45 (0.29%) | 2 (0.49%) | 11 (0.75%) |
| Laparoscopic approach | 2177 (94.24%) | 14,519 (94.92%) | 343 (84.48%) | 1243 (85.02%) |
| Total | 2310 | 15,295 | 406 | 1462 |
HH: hiatal hernia
Within the elective cohort, octogenarians were at increased risk for mortality (ARD 3.6%, 95% CI 2.7–4.5%), malnutrition (ARD 3.6%, 95% CI 2.6–4.7%), sepsis (ARD 1.9%, 95% CI 1.1–2.7%), respiratory failure (ARD 3.3%, 95% CI 2.3–4.2%), pneumonia (ARD 2.8%, 95% CI 1.8–3.8%), DVT (ARD 1.4%, 95% CI 0.6–2.1%), blood transfusion (ARD 0.6%, 95% CI 0.1–1%), and discharge to a nursing facility (ARD 0.9%, 95% CI 0.3–1.5%) (Table 3). Within the emergent cohort, octogenarians were at increased risk for mortality (ARD 6.7%, 95% CI 3.6–9.7%), malnutrition (ARD 6.3%, 95% CI 2.8–9.7%), sepsis (ARD 4.7%, 95% CI 1.7–7.8%), and respiratory failure (ARD 5.9%, 95% CI 2.5–9.2%) (Table 4).
Table 3.
90 day outcomes in the elective repair cohort
| Elective HH | |||
|---|---|---|---|
| n (%) | |||
| Octogenarian | Senior | Absolute risk difference (95% CI) | |
| n | 2310 | 15,295 | |
| Mortality | 109 (4.7%) | 169 (1.1%) | 3.6% (2.7%, 4.5%) |
| Reoperation | 12 (0.5%) | 71 (0.5%) | 0.1% (− 0.3%, 0.4%) |
| Malnutrition | 146 (6.3%) | 409 (2.7%) | 3.6% (2.6%, 4.7%) |
| Sepsis | 78 (3.4%) | 223 (1.5%) | 1.9% (1.1%, 2.7%) |
| Respiratory failure | 127 (5.5%) | 343 (2.2%) | 3.3% (2.3%, 4.2%) |
| Pneumonia | 119 (5.2%) | 360 (2.4%) | 2.8% (1.8%, 3.8%) |
| DVT | 66 (2.9%) | 226 (1.5%) | 1.4% (0.6%, 2.1%) |
| PE | 41 (1.8%) | 261 (1.7%) | 0.1% (− 0.5%, 0.7%) |
| Blood transfusion | 27 (1.2%) | 93 (0.6%) | 0.6% (0.1%, 1%) |
| Need for Physical therapy | 98 (4.2%) | 645 (4.2%) | < 0.0% (− 0.9%, 0.9%) |
| Discharge to nursing facility | 41 (1.8%) | 129 (0.8%) | 0.9% (0.3%, 1.5%) |
Table 4.
90 day outcomes in the emergent repair cohort
| Emergent HH | |||
|---|---|---|---|
| n (%) | |||
| Octogenarian | Senior | Absolute risk difference (95% CI) | |
| n | 406 | 1462 | |
| Mortality | 37 (9.1%) | 36 (2.5%) | 6.7% (3.6%, 9.7%) |
| Reoperation | 2 (0.5%) | 11 (0.8%) | − 0.3% (− 1.2%, 0.7%) |
| Malnutrition | 46 (11.3%) | 74 (5.1%) | 6.3% (2.8%, 9.7%) |
| Sepsis | 35 (8.6%) | 57 (3.9%) | 4.7% (1.7%, 7.8%) |
| Respiratory failure | 43 (10.6%) | 69 (4.7%) | 5.9% (2.5%, 9.2%) |
| Pneumonia | 29 (7.1%) | 73 (5.0%) | 2.1% (− 0.8%, 5.1%) |
| DVT | 18 (4.4%) | 44 (3.0%) | 1.4% (− 0.9%, 3.8%) |
| PE | 10 (2.5%) | 39 (2.7%) | − 0.2% (− 2.0%, 1.7%) |
| Blood transfusion | 6 (1.5%) | 19 (1.3%) | 0.2% (− 1.3%, 1.6%) |
| Need for Physical therapy | 22 (5.5%) | 77 (5.3%) | 0.2% (− 2.5%, 2.8%) |
| Discharge to nursing facility | 11 (2.7%) | 34 (2.3%) | 0.4% (− 1.5%, 2.3%) |
Utilizing the logistic regression models, the OR for mortality in octogenarians as compared to seniors was 3.9 (95% CI 3.1–5.0) in elective cases and 3.5 (95% CI 2.1–5.6) in emergent cases. The OR for respiratory failure was 2.4 (95% CI 1.9–3.0) for elective cases and 2.1 (95% CI 1.4–3.2) for emergent cases. The OR for sepsis was 2.2 (95% CI 1.7–2.9) in elective cases and 2.1 (95% CI 1.4–3.4) for emergent cases. These results are presented in Figs. 2, 3.
Fig. 2.
Odds ratios of octogenarians vs seniors for primary outcomes in the elective cohort
Fig. 3.
Odds ratios of octogenarians versus seniors for primary outcomes in the emergent cohort
Discussion
This retrospective study utilizing the TriNetX database found that octogenarians undergoing both elective and emergent hiatal hernia repair are at significantly increased risk for death and other complications compared to patients aged 65–79. These effects persisted after controlling for meaningful differences in comorbidities between the two groups. This suggests that senior-age patients who delay elective repair of hiatal hernia until the 8th decade of life are exposing themselves to significantly elevated risk for poor outcomes.
Wilson et. al have recently published data with a similar design, using the Nationwide Readmission Database to compare octogenarians against adult patients aged 18–79 undergoing paraesophageal hernia repair [10]. After controlling for multiple variables including Charlson Comorbidity Index scores, they did not find that octogenarian status was interpedently predictive of mortality, though undergoing an emergent operation was. This is somewhat surprising given the much younger cohort they were comparing against octogenarians but may be related to the number of variables controlled for in their models.
Single institution data has previously demonstrated that both age over 72 and high frailty double the odds of morbidity after elective repair of paraesophageal hernia [11]. This study replicates a similar finding on the effects of age in a national database. Age and frailty have similarly been found to be meaningful predictors of mortality after emergent repair in large, national databases [12]. These findings are consistent with the current study, even though frailty was not directly assessed, as the variables controlled for in the multivariate model have some overlap with the modified frailty index.
The Stylopoulos paper has been followed by two similar papers utilizing relatively more modern data [8, 13, 14]. At the time of this manuscript preparation, these three papers are important citations in the proposed update to the SAGES Guideline for Hiatal Hernia Repair which is currently available for public comment [15]. Interestingly, all three of these models assume a mortality of 5.4–5.5% in emergent repairs while the mortality rate across patients age 65–89 in this cohort was 3.9%.
The data on the progression from asymptomatic hiatal hernias to symptomatic or emergent is very limited. In light of the increased utilization of minimally invasive surgery, its increased safety profile, and the reported mortality rates, the Authors would encourage further study. As data accumulates, a reevaluation of recommendations addressing whether to operate on the asymptomatic hiatal hernia in the senior or geriatric population may be warranted. A European expert Delphi consensus published in 2023 sought to answer the question of indications for surgical repair of paraesophageal hernia and may signal a change in surgeons’ perspective [16]. They labeled surgery the “recommended” therapeutic strategy for symptomatic patients regardless of age and considered surgery “acceptable” for patients with no or minor symptoms.
This study has a number of limitations. Type I, or sliding, hiatal hernias were likely included so it is not truly representative of paraesophageal hernias alone. However this is unlikely to minimize the demonstrated effects. This study also had no ability to distinguish the operation each patient received.
Avenues for future research include further differentiating the effects of age versus frailty on the outcomes of elective hiatal hernia repair. The models investigating the outcomes of elective hiatal hernia repair versus observation may also benefit from being updated with more modern data.
The results of this study strongly suggest that senior-age patients considering hiatal hernia repair would have significantly better outcomes with immediate elective repair rather than pursuing repair after age 80.
Supplementary Information
Below is the link to the electronic supplementary material.
Funding
We have no external sources of funding to report.
Declarations
Disclosures
Sunjay S. Kumar, Martina Rama, Scott Koeneman, Sami Tannouri, Talar Tatarian, and Francesco Palazzo have no disclosures.
Footnotes
This project was presented at the 2024 SAGES Annual Meeting in Cleveland, Ohio, USA on April 17, 2024.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Till BM, Mack SJ, Whitehorn G, Micaela LC, Chi-Fu JY, Grenda T, Evans NR, Okusanya O. The epidemiology of surgically managed hiatal hernia: a nine year review of national trends. Foregut. 2023;3:20–28. doi: 10.1177/26345161221137089. [DOI] [Google Scholar]
- 2.Lou J, Kooragayala K, Williams J, Kalola A, Crudeli C, Sandilos G, Butchy MV, Shersher DD, Burg JM. Diagnostic workup and therapeutic intervention of hiatal hernias discovered as incidental findings on computed tomography. Am Surg. 2024 doi: 10.1177/00031348241230096. [DOI] [PubMed] [Google Scholar]
- 3.Caplan Z (2023) U.S. Older population grew from 2010 to 2020 at fastest rate since 1880 to 1890. census.gov
- 4.Soresi AL. Diaphragmatic hernia: its unsuspected frequency: its diagnosis: technic for radical cure. Ann Surg. 1919;69:254–270. doi: 10.1097/00000658-191903000-00005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Brindley GV. Surgical treatment of complications of esopliageal hiatal hernia. Ann Surg. 1965;161:723–731. doi: 10.1097/00000658-196505000-00010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Skinner DB, Belsey RH. Surgical management of esophageal reflux and hiatus hernia. Long-term results with 1,030 patients. J Thorac Cardiovasc Surg. 1967;53:33–54. doi: 10.1016/S0022-5223(19)43239-X. [DOI] [PubMed] [Google Scholar]
- 7.Kohn GP, Price RR, DeMeester SR, Zehetner J, Muensterer OJ, Awad Z, Mittal SK, Richardson WS, Stefanidis D, Fanelli RD. Guidelines for the management of hiatal hernia. Surg Endosc. 2013;27:4409–4428. doi: 10.1007/s00464-013-3173-3. [DOI] [PubMed] [Google Scholar]
- 8.Stylopoulos N, Gazelle GS, Rattner DW. Paraesophageal hernias: operation or observation? Ann Surg. 2002;236:492–501. doi: 10.1097/00000658-200210000-00012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Schlottmann F, Strassle PD, Allaix ME, Patti MG. Paraesophageal hernia repair in the USA: trends of utilization stratified by surgical volume and consequent impact on perioperative outcomes. J Gastrointest Surg. 2017;21:1199–1205. doi: 10.1007/s11605-017-3469-z. [DOI] [PubMed] [Google Scholar]
- 10.Wilson HH, Ayuso SA, Rose M, Ku D, Scarola GT, Augenstein VA, Colavita PD, Heniford BT. Defining surgical risk in octogenarians undergoing paraesophageal hernia repair. Surg Endosc. 2023;37:8644–8654. doi: 10.1007/s00464-023-10270-z. [DOI] [PubMed] [Google Scholar]
- 11.Wang TN, An BW, Wang TX, McNamara M, Sweigert PJ, Yuce TK, Heh V, Collins CE, Haisley KR, Perry KA. The effect of frailty and age on outcomes in elective paraesophageal hernia repair. Surg Endosc. 2023;37:9514–9522. doi: 10.1007/s00464-023-10363-9. [DOI] [PubMed] [Google Scholar]
- 12.Collins ML, Mack SJ, Till BM, Whitehorn GL, Tofani C, Chojnacki K, Grenda T, Evans NR, Okusanya OT. Defining risk factors for mortality after emergent hiatal hernia repair in the era of minimally invasive surgery. Am J Surg. 2023;225:1056–1061. doi: 10.1016/j.amjsurg.2023.01.012. [DOI] [PubMed] [Google Scholar]
- 13.Morrow EH, Chen J, Patel R, Bellows B, Nirula R, Glasgow R, Nelson RE. Watchful waiting versus elective repair for asymptomatic and minimally symptomatic paraesophageal hernias: a cost-effectiveness analysis. Am J Surg. 2018;216:760–763. doi: 10.1016/j.amjsurg.2018.07.037. [DOI] [PubMed] [Google Scholar]
- 14.Jung JJ, Naimark DM, Behman R, Grantcharov TP. Approach to asymptomatic paraesophageal hernia: watchful waiting or elective laparoscopic hernia repair? Surg Endosc. 2018;32:864–871. doi: 10.1007/s00464-017-5755-y. [DOI] [PubMed] [Google Scholar]
- 15.Daly S, Kumar SS, Collings AT, Hanna NM, Pandya YK, Kurtz J, Kooragayala K, Barber MW, Paranyak M, Kurian M, Chiu J, Ansari MT, Slater BJ, Kohn G (2024) Proposed-SAGES guidelines for the surgical treatment of hiatal hernias. sages.org
- 16.Gerdes S, Schoppmann SF, Bonavina L, Boyle N, Müller-Stich BP, Gutschow CA, Hiatus Hernia Delphi Collaborative Group Management of paraesophageal hiatus hernia: recommendations following a European expert Delphi consensus. Surg Endosc. 2023;37:4555–4565. doi: 10.1007/s00464-023-09933-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.