Mortality and complications after emergency laparotomy in patients above 80 years

Emergency laparotomy (EL) is a high‐risk procedure. However, available evidence regarding outcome after emergency surgery in very old patients is limited. The aim of this observational study was to investigate outcome following EL in patients ≥80 years of age.


| BACKG ROU N D
Emergency laparotomy is a high-risk procedure in old patients due to underlying comorbidity, frailty, and reduced physiological reserves. 1 Previously, there has been little focus on emergency abdominal surgery patients, but this is changing, as the National Emergency Laparotomy Audit (NELA) is expanding in the United Kingdom. 2 In Denmark, a national clinical database of emergency abdominal surgery is being established, collecting data regarding clinical outcome after surgery for bowel obstruction and perforated hollow viscus. 3 The first report from the NELA initiative 4 revealed huge variations in standards of care after EL and thus several authors now advocate care bundles in emergency abdominal surgery. 5,6 However, many Nordic hospitals have not implemented such standardized pathways yet.
Emergency laparotomy is performed more often in older than in younger patients. 6 Nevertheless, available evidence regarding emergency surgery in octo-and nonagenarians is limited, and few studies have addressed outcome after EL in this group. [7][8][9][10][11][12][13][14] The oldest age groups are now expanding rapidly, and the number of octogenarians in Norway is expected to double within 20 years. 15 Therefore, the aim of this study was to investigate mortality and morbidity in patients aged ≥80 years undergoing emergency laparotomy in a Norwegian university hospital without a standardized care bundle for emergency laparotomy.

| ME THODS
This is a single-center retrospective cohort study. In April 2017, the electronic operation planning system at Haukeland University Hospital, Norway, was queried for patients ≥80 years old undergoing emergency laparotomy during the period January 2015 through December 2016. Exclusion criteria were pure palliative surgery and emergency vascular surgery. The primary outcome was 30-day

Editorial Comment:
In this observational single-center study, findings of high morbidity, mortality and a substantial functional decline are reported after emergency laparotomy in patients above 80 years. Structural changes in the treatment program for these older patients are suggested.
F I G U R E 1 Kaplan-Meier curve for 1-year survival after emergency laparotomy, with 95% pointwise confidence intervals (n = 106) [Colour figure can be viewed at wileyonlinelibrary.com] mortality, and secondary outcomes were 90-day mortality, 1-year mortality, in-hospital complications, and level of care at discharge. Data regarding age, sex, American Society of Anesthesiologists physical status classification (ASA), comorbidities, residential status, and number of daily medications were retrieved from the medical records. Daily medication was defined as any medication prescribed by a physician, also including vitamin substitution and herbal medication. The following parameters were recorded: duration from CT scan until start of surgery, the operative procedures performed, in-  18 Complications with Clavien-Dindo grade ≥2 were included.
To identify delirium, a chart-based method described by Inouye was used. 19 The project was approved by the Regional Ethics Committee, who waived informed consent (REK 2017/610).

| Statistical analysis
We present categorical data as counts and percentages (with 95% confidence intervals), and continuous data as means, medians and/ or ranges, as appropriate. Confidence intervals for percentages were calculated using the recommended Wilson (score) method. 20 Survival up to 1 year is also shown as a Kaplan-Meier survival curve ( Figure 1), with 95% pointwise confidence bands. The 30-day standardized mortality ratio (SMR) was calculated using age-and sexmatched mortality data for the general Norwegian population. 21 The amount of missing data was low (<5% for all variables), so for variables with missing data, we used complete-case analysis. All statistical analyses were done using SPSS 24 (Chicago IL USA) and/or R version 3.6.0. 22

| RE SULTS
All eligible patients undergoing emergency laparotomy during the 2-year period were included, in total 106 patients. Demographic data regarding age, sex, ASA classification, and comorbidities are shown in Table 1. Comorbidity was present in 102 patients (96%), and 62 (58%) had cardiopulmonary disease, that is, COPD, emphysema, asthma, cardiac arrhythmia, ischemic heart disease, or congestive heart failure. Cancer other than gastrointestinal was present in 11 (10%) patients. The median number of daily prescriptions in the cohort was 5.A table demonstrating the 20 most common classes of medications taken by the patients is shown in "Supplementary material" (Table S1). Pre-operatively, 17 (16%) patients were nursing home residents, and 89 (84%) lived at home.
The operative procedures, intraoperative findings, and intrahospital logistics are described in Table 2. Resection of colon was performed in 25%, followed by adhesiolysis, and resection of small intestine. Median time from CT-scan until surgery was 9.0 hours, median stay in PACU was 18 hours, and 12 patients (11%) were treated in the ICU post-operatively.
Post-operative complications are shown in Figure 2 and Table S2.
Mortality and discharge data are shown in Table 3. For 82% of the patients, at least one complication occurred. Pulmonary complications (POPCs) were the most frequent complication (48%), followed by delirium (40%). Kidney failure and paralytic ileus were equally common (22%), and wound infection occurred in 17%. The number of intrahospital deaths was 25 (24%), and of the 81 surviving patients, 53 were discharged to a nursing home post-operatively. The median length of hospital stay was 9 days (range 0.6-50 days). The total 30-day, 90-day, and 1-year mortality were 26%, 34%, and 47%, respectively.  In the cohort, there were 82 patients without cancer, and their 1-year mortality rate was 44%. The 1-year mortality rate among the 24 patients with either previously known cancer (n = 10) or cancer diagnosed at surgery (n = 14) was 58%. Developing a complication is a serious event, and earlier work has shown that the risk of dying could increase threefold. 8 In our study, Similar findings are commonly reported after abdominal surgery in the elderly, and up to a threefold increase in pulmonary complications has been reported in patients ≥85 years. 28 The second most common morbidity was delirium, which was observed in 40% of the patients. According to the formal definition, 17 this complication should only be registered when based on strict criteria. Use of a delirium score is not common practice in Norwegian surgical wards. Hence, we used a chart-based method for evaluating delirium in the hospitalized patients retrospectively. 19 The frequency of delirium is probably underestimated, as hypoactive delirium is difficult to recognize, and some of the cognitive problems could represent post-operative cognitive dysfunction. A high prevalence of cognitive dysfunction has also been reported in other studies in patients hospitalized with acute surgical morbidity. One study found delirium in 2/3 of elderly patients admitted for general surgical conditions, 29 while a retrospective study reported a 33%

| D ISCUSS I ON
incidence of post-operative delirium after elective major surgery. 30 The evidence regarding functional decline or discharge to institution after emergency laparotomy is scarce. [12][13][14] However, in one study, 60% of the patients needed assistance with ADL, 13 and in another paper, 55% of patients were discharged to long-time care. 12 Both results are similar to our findings.
The majority of our patients were vulnerable, as only four patients were admitted without any previous comorbidity. This is comparable to data from other studies, [7][8][9][10][11][12][13][14] where comorbidities were present in 75%-90% of the patients. and this contributes to high morbidity and mortality. 4 Our study demonstrated a wide range in duration between CT scan and surgery (0.5-156 hours). The delay could be due to the fact that CT scans sometimes are inconclusive, indicating a chance that the patient might recover spontaneously. In such case, a strategy of "watch and wait" is appropriate. However, long interval between diagnose and treatment could also reflect scarce capacity or low priority in the OR.
During The study has several limitations. It is a retrospective single-center study, limiting external generalization of the data, and with a small number of patients included. However, our results are in accordance with reports from larger populations.
Due to lack of information, we did not score frailty retrospectively.
The modified frailty index (MFI) has been used in large studies, 32 but this score measures comorbidity rather than frailty. 33 We suspect, however, that a large proportion of our patients were frail, as frailty is reported in 28%-79.9% of elderly emergency surgery patients. 34,35 The number of patients deemed unfit and thus rejected for surgery in this 2-year period is unknown, but suspected to be very low, as offering surgery when indicated, even if the patient is very fragile, has been common practice in our hospital.
In conclusion, our study demonstrates that poor outcome after emergency laparotomy in octo-and nonagenarians is common, with high mortality and morbidity, and hence a high risk of functional

ACK N OWLED G EM ENTS
The authors thank MD, Professor Hans Flaatten and MD, PhD Kristin Moberg Aakre for their contributions revising this manuscript. The study was performed without external funding.

CO N FLI C T O F I NTE R E S T
All authors declare that they have no conflict of interest.