Postoperative Outcome and Predictors of Mortality in Octogenarians Patients with Gastric Cancer

Postoperative Outcome and Predictors of Mortality in Octogenarians Patients Gastric Cancer. Abstract Background: Gastric cancer is most frequent after the fifth decade of life. Surgical risk is higher in aged population be-cause of general health condition may affect the postoperative result. Aim of the study was to identify risk factors for post-operative mortality in octogenarian patients who underwent surgery for gastric cancer. Methods: 236 patients (181: 80-85 years old and 55: >85 years old) underwent surgery for gastric cancer at the Sant’Orsola-Malpighi University Hospital in Bologna between 2012 and 2017. The variables of the two groups of pa-tients were compared. Results: Post-operative mortality was 5.5% among 80-85 years old and 9% for the > 85 years old. The two groups sig-nificantly differed in: age (p< 0.0001), type of hospitalization (p = 0.005), site of primary tumor (pylorus p =0.030; body p = 0.001), presence of cardiac comorbidities (p = 0.043), ASA score (p = 0.021) and type of surgical presentation (elective vs urgent surgery, p = 0.001). In the multivariate analysis, urgent surgery (p= 0.002) and ASA score >III (p = 0.021) were associated to post-operative mortality; ASA score (p = 0.041) and stage (p = 0.012) significantly influ-enced survival. Conclusion: ASA score>III and urgent surgery, but not age ≥ 85 years, were associated with postoperative mortality. In the very elderly, alias oldest old patients, preoperative nutritional status and pre-existing comorbidities, rather than age itself, should be considered as selection criteria for surgery. ASA risk assessment may be beneficial for stratification of patients and for ultimately optimizing outcomes.


Introduction
Gastric cancer is the fifth most common malignancy in men and the second in women, with 952,000 new cases diag-nosed per year worldwide [1]. In Italy gastric cancer represent the fifth most common malignancy in the men and the sixth in women, with 14.220 new cases diagnosed in the 2013 year [2]. The treatment of choice for gastric cancer is surgical resection and is both curative and preventive of gastric cancer re-lated complications. Gastric cancer

Inclusion Criteria
A total of 236 patients aged 80 years or older with histologically confirmed primary gastric cancer underwent surgery in Sant'Orsola-Malpighi Hospital in Bologna between 2012 and 2017.
The sample included elective and urgent surgery (both resections and palliative surgery). Clinico-pathological data for these patients were obtained from hospital rec-ords. Comprehensive informed consent was obtained from all patients when they admitted our hospital prior to sur-gery.

Definition and Classification of Variables
We divided the patients into two categories according to the age: from 80 to 85 years or >85 years. The site of primary tumor was categorized as "cardias," "pylorus," "body," "bottom", "not specified or "other". The staging of the tumor was assigned according to the TNM (Tumor Node Metastases) classification of the American Joint Committee on Cancer/Union for International Cancer Control [6,7]. We recorded the ASA score referring to the classification of the American Society of Anesthesiologists [8].
The type of surgical presentation was categorized as "elective" or "urgent." Surgical operation was classified as "partial gastrectomy with jejunal anastomosis," "total gastrectomy," "gastroenter-ostomy without gastrectomy" or "other" (palliative surgery). Surgical approach was categorized as "Roux," "Billroth II" or "other." Postoperative complications were defined as events that occurred during hospitalization following the operation and were evaluated according to Clavien-Dindo classification. Clavien score was referred to Clavien-Dindo classification of surgical complications [9]. Length of hospitalization was defined as the number of days from the date of surgical intervention to discharge. Post-operative mortality was defined that occurred within 30 days from surgery. In-hospital mortality was defined that occurred during hospitalization up to 90 days excluded post-operative mortality. Overall survival (OS) was defined as the time from the date of surgery to patient death (including surgery-associated death or hospital death), or the date of last available information concerning vital status.
Tumor location, clinical or pathological stage, degree of lymph node dissection (D0, <D2 or >D2), and curability were assessed according to the Japanese Classification of Gastric Carcinoma, 13 th , and then 14 th editions [6,10,11]. Surgical mortality, morbidity, and hospital mortality were compared between two groups.
Recurrences were confirmed by computed tomography, tumor markers, and endoscopic examinations.

Statistical Analysis
The sample was divided into two cohorts: patients from 80 to 85 and >85. Data are represented as median and inter-quartile range (IQR) for continuous variables and as n (%) for categorical variables.

Characteristics of the Patients
The following tables show a comparison of the variables in detail (Tables 1-4     as independent prognostic factors that significantly influenced survival.  In our report, the rate of postoperative mortality in >85 years old was significantly higher than in 80-85 years old (9% versus 5.5%). Multivariate analysis of the combined cohort indicated however that the age>85 did not correlate with mortality; on the contrary, urgent surgical presentation and an ASA score>III were associated with postoperative mor-tality and with one-year surgery survival. In the combined cohort, ASA>III and urgent surgery were associated with postoperative mortality. In literature, there is wide consensus on the causal relation "ASA -post-operative mortality" for elderly patients who have undergone gastric cancer surgery [12,13]. According to our experience, in the oldest old, the higher ASA score may correlate to lower functional reserves and more significant comorbidities, enough to influence short-term outcomes.
The correlation between urgent surgical presentation and operative mortality, for the oldest old undergoing gastric can-cer surgery, is confirmed in literature [14]. The hypothesis is that bowel obstruction or perforation may carry a poor prognosis independently from the surgical operation; it is possible that the surgical procedure physical and mental stress along with the general anaesthesia further influence short-term outcomes in such frail conditions. According to several observations, important predictors of postoperative mortality in oldest old patients who have un-dergone surgery for gastric cancer are ASA score and urgent surgical presentation [15]. Age did not represent a factor associated with postoperative mortality in previous reports [16][17][18].
Considering these results, it appears appropriate planning prospective studies in order to achieve even stronger evidence regarding the safety of the oldest old undergo-ing gastric cancer surgery. Age did not represent an independent risk factor for postoperative mortality; patients should be assessed according to their general physical condition, whatever the age. In several studies similar to our own, ASA score and albumin serum concentration are reported as indirect indicators of the patient's functional conditions [19][20][21]. In recent research, multidimensional geriatric assessment has shown to be useful in giving a 360° picture of a patient's status, in terms of physical and functional condition [22,23] This study has the limitations of retrospective investigations involving chart review. Moreover, given the rarity of an operation on a nonagenarian, the number of the sample might limit the statistical power of and limit the generalization of the conclusions. A further limitation of our report was the lack of a standardized institutional protocol for postopera-tive care.

Conclusion
From the current analysis of a large database of 236 over octogenarians' patients who have undergone gastric cancer surgery, urgent surgical presentation and ASA score>III, but not age≥85, were correlated to postoperative mortality. These results should prompt an improvement in multidimensional geriatric assessment, to better define the risk stratifi-cation in the oldest old patients undergoing gastric surgery. Moreover, the development of effective corrective interven-tions of the above preoperative functional variables might lead to an improvement of surgical outcomes in the increas-ingly large geriatric population.

Compliance with Ethical Guidelines
Conflict of interest Lugaresi M, Duchi A, Di Saverio S, Argento F, Yavuzdiler KE, Martini F, Novello M, Cavallari G, Nar-do B declare that they have no competing interests.