Since centralization of Upper Gastro-Intestinal cancer services in 2009, exponential growth in referrals for oesophago-gastric emergencies such as esophageal perforations and incarcerated giant hiatus hernias (GHH) was noted. This tertiary referral unit maintains prospective data of all the giant hernias (>30% stomach in chest), that undergo elective as well as emergency surgery.

GHH comprise 5–10% of all hiatus hernias (1) and mainly affects elderly population (2, 3). Current guidelines recommend elective surgery for symptomatic hernias and watchful wait for asymptomatic ones (4, 5). Acute complications requiring emergency repair include gastric volvulus, uncontrolled bleeding, obstruction, strangulation, and perforation (2, 6). The risk of developing acute symptoms that necessitate emergency surgery is estimated between 0.7 and 7% (7). Despite this relatively low incidence, clinicians must be mindful of these patients, as the mortality rate of emergency surgery ranges from 7 to 50%, which is much higher than that of elective repair (0.65%) (8, 9). Patients who require an emergency operation are also subject to 5-fold increased odds of serious morbidity compared with patients undergoing elective surgery (8).

Aim of this study is to present the short-term outcomes of non-elective GHH repair in a single institution over a 10-year period, review the pertinent literature and provide insights into decisions for asymptomatic patients.

A total of 42 consecutive patients underwent emergency or urgent repair of GHH between January 2009 and October 2019. Of these, five patients were excluded because their medical records were not available for review. The demographics, baseline clinical characteristics and presenting symptoms of the included patients are summarized in Table 1. Transfers from other hospitals accounted for 46% of admissions. Diagnostic investigations included computed tomography (CT) in 34 patients (91.9%), oesophagogastroduodenoscopy (OGD) in 11 (29.7%) and barium swallow imaging in 3 (8.1%). The surgical management of the patients is outlined in Table 2.

The surgical outcomes are detailed in Table 3. The commonest postoperative complication was pneumonia, occurring in 13 patients (35.1%). Seven patients (18.9%) required total parenteral nutrition, 3 (8.1%) developed pneumothorax that necessitated chest drain insertion, and 2 patients (5.4%) had new-onset atrial fibrillation that was treated with correction of electrolyte disturbances. There were no recorded cases of postoperative myocardial infarction, pulmonary embolism, or cerebrovascular accident.

Two deaths occurred in the patient cohort. The first patient was an 86-year-old man with a Charlson comorbidity index of 7 who was transferred from another hospital with sepsis and symptoms of gastrointestinal obstruction. He underwent open repair and died on the fourth postoperative day from multiorgan failure. The second patient was an 84-year-old man with a Charlson comorbidity index of 6 who presented with coffee ground vomiting and was transferred from another hospital. Eleven days following open repair, he developed peritonitis secondary to perforation of the transverse colon and underwent a further open repair. He died 7 days later.

Two patients who had very poor functional outcomes despite adequate restoration of anatomy and are discussed here.

A 51-year-old man presented with epigastric pain and vomiting. On the third day of admission, he underwent laparoscopic repair of a herniated gastric fundus through a large crural defect. He subsequently developed progressive dysphagia, and a barium swallow investigation revealed a long esophageal stricture which was considered to be of peptic etiology. This required multiple endoscopic balloon dilations. A follow-up barium swallow also showed narrowing at the level of the pylorus, suggestive of gastric outlet obstruction. He has been monitored with regular outpatient clinic appointments.

A 78-year-old lady presented with profuse coffee-ground vomiting secondary to GHH and underwent laparoscopic repair on the fourth day of admission. She was found to have almost complete herniation of the stomach with partial volvulus, and a gastrostomy tube was inserted. Postoperatively, she developed intractable vomiting. Investigations revealed no evidence of gastric outlet obstruction but there was significant delay in gastric emptying, suggestive of gastroparesis. As no clinical improvement was noted, a decision was made to use a jejunal extension of the gastrostomy to maintain nutritional requirements. The patient died 3 years later from an unrelated cause.

Whereas long-term outcomes in emergency GHH repairs have been thoroughly discussed, short-term outcomes have been less well studied. Compared to the available literature, the recurrence rate was low (5.4%, n = 2) and mortality was acceptable (5.4%, n = 2). Mortality for acute presentations is higher when compared to our previous published data for elective laparoscopic repairs of GHH (2.9%) (15, 16). However, morbidity was high (35%) and some of the complications were unusual and difficult to explain.

Probably due to the advanced age, most of our patients had comorbid conditions, with more than half suffering from a cardiorespiratory disease. Similar were the findings in other studies, in which patients presenting acutely were older and had higher prevalence of at least one cardiopulmonary comorbidity compared to those undergoing elective GHH repair (17–19). In our study, the median Charlson comorbidity index score was 3 (range, 0–7). Tam et al. (17) reported an age-adjusted Charlson comorbidity index score of 4 (range, 0–6) for 171 patients undergoing non-elective repair of paraesophageal hernia. In contrast, the same score was significantly lower (2; range, 0–4; p < 0.0001) for 753 patients who underwent elective repair.

Advanced age and comorbidity, along with urgency of surgery, have all been postulated to predict patient-specific risk of postoperative morbidity and mortality. Ballian et al. (20) assessed 980 patients who underwent GHH repair (80% elective and 97% laparoscopic) and developed clinical prediction models for in-hospital or 30-day mortality and major morbidity using logistic regression. A 4-covariate prediction model consisting of age >80 years, pulmonary disease, congestive heart failure and urgency of operation provided discriminatory accuracy for postoperative mortality of 88%. A 5-covariate model, which further included sex, was 68% predictive of major postoperative morbidity. Urgency of operation was a significant predictor of major morbidity (non-elective 41% vs. elective 18%; P < 0.001) and mortality (non-elective 8% vs. elective 1.1%; P < 0.001). Markar et al. (21) investigated the influence of hospital volume upon clinical outcomes by retrospectively reviewing 12,441 cases of paraesophageal hernia causing obstruction or gangrene. High hospital volume was an independent predictor of reduced 30-day mortality (odds ratio 0.65, 95% confidence interval 0.55–0.76, P < 0.001). In addition, multivariate regression analysis showed significant factors associated with 30-day mortality included age >70 years, chronic congestive cardiac failure, ischemic heart disease, dementia, as well as pulmonary and renal disease. In this study, the logistic regression model that was performed to ascertain the effects of age, ASA class and time to surgery on major morbidity was not statistically significant (P = 0.18).

Several studies published over the last two decades have reported on surgical outcomes following non-elective repair of GHH. Table 4 presents the results of such studies, with the exclusion of registry-based studies, those that did not report on emergency data separately and studies with <10 patients. Complication rates varied significantly, ranging from 28 to 100%. Major postoperative morbidity (Clavien-Dindo grades III or IV complications) ranged from 8 to 30%. In our study, this rate was slightly higher at 35%. Recurrence rates ranged between 7 and 18%. In the present study, recurrence occurred only in 2 patients (5.4%), both of which underwent revision surgery. The mortality of emergency paraesophageal hernia repair varied between 0 and 22%.

In our unit we also follow-up patients with GHH who either decline surgery or are deemed unsuitable for operative management. Follow-up is augmented by the Quality of Life in Reflux and Dyspepsia (QOLRAD) questionnaire for the first 2 years. Since 2014, a total of 17 such patients were followed up. Of these, four decided to have surgery at a later date and only one presented acutely with volvulus. Longer follow-up times might reveal a higher number of patients who ultimately need surgery.

The main limitations of the present study include the retrospective collection of data and relatively small number of patients, probably because of rarity of condition itself. Furthermore, unlike the standardized surgical approach in elective surgery, there is variability in the emergency situation. While operating on high risk, elderly and frail patients it may be necessary to adopt a minimalistic approach and only perform gastropexy or gastrostomy to prevent volvulus. In our unit we attempt complete excision of the sac followed by crural repair and fundoplication. However, this is not always possible in high-risk patients leading to heterogeneity in surgical technique.

Patients admitted with acute GHH symptoms are older and have more comorbid conditions compared to their counterparts who undergo elective surgical repair. Major postoperative complications, which require intervention, repeat surgery or admission to the intensive care unit, may occur in as many as one in three patients. Proceeding to surgery should be thoroughly discussed with these patients as surgery may save their life but leave them with major morbidity. Operating on these patients in the elective setting might avoid high morbidity and mortality. Larger studies to investigate prediction models for post-operative complications will allow risk stratification and informed management decisions.

The original contributions presented in the study are included in the article/supplementary materials, further inquiries can be directed to the corresponding author.

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

IB: literature review and writing up of article. DA: data collection. SL: writing up of article. RD: concept, supervision, and review of final article. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.