Abstract
Background/Aim: The aim of this study was to determine the significance of immunonutritional and physical index in the assessment of risk associated with pancreaticoduodenectomy (PD) in the elderly. Patients and Methods: This study enrolled 92 patients who underwent PD. They were divided into 2 groups: Group A included patients 79 years and younger (n=79) and Group B patients 80 years and older (n=13). Among 37 patients, physical function and body composition were also evaluated. Results: Significantly higher neutrophil–lymphocyte ratio, lower prognostic nutritional index (PNI), and controlling nutritional score were observed in Group B. Muscle strength and walking ability were significantly impaired in Group B, although there was no significant difference in body composition. Age was not correlated with the incidence of postoperative complications, overall survival or recurrence-free survival by univariate and multivariate analysis. Conclusion: PD is justified for the elderly, with acceptable morbidity and prognosis. However, immunonutritional status and physical function are significantly impaired; thus, appropriate case selection and active nutritional support are required for the elderly.
Pancreaticoduodenectomy (PD) is the only curative option for pancreatic head, distal biliary tract, and duodenal malignant tumors. The number of elderly patients who undergo PD is rising in Japan, where the elderly population is rapidly increasing as in many other countries. These patients are generally and potentially characterized by declining physiological functions, poor nutritional status, impaired physical activity and cognitive function. Elderly patients may have poorer prognoses compared to younger patients because of their increased risk of postoperative complications and fewer expected life years.
Perioperative management of patients undergoing PD has dramatically improved over the past few decades, but the morbidity rate still ranges from 20% to 50% (1, 2). In the past, dismal early and long-term outcomes restricted PD to younger patients who did not have comorbid diseases (3). A recent cohort study reported that PD in elderly patients was still associated with significantly more postoperative complications and higher costs (4). However, on the contrary, a systematic literature review of 21 studies reported that PD was feasible in elderly patients, with acceptable morbidity and mortality rates (5). However, because of a high incidence of comorbid diseases, elderly patients are still generally considered high-risk patients for major surgery. Given these conflicting data, objective indicators for choosing the surgical indication for the elderly are strongly needed.
Preoperative malnutrition and sarcopenia are associated with a higher incidence of complications after PD (6-8), and preoperative administration of an immune-enhancing diet is reported to improve clinical outcomes in patients undergoing PD (9). Aging as well as underlying malignancy are associated with malnutrition and sarcopenia; therefore, assessment of the nutritional condition is critical to decide whether elderly patients should undergo surgery. Perioperative nutritional support and preoperative rehabilitation are beneficial for elderly patients who will undergo major abdominal surgery, and these practices can reduce the rates of surgical complications and mortality (10, 11). A number of indicators, including the neutrophil–to–lymphocyte ratio (NLR), the prognostic nutritional index (PNI), the Glasgow Prognostic Score (GPS), and the Controlling Nutritional Status (CONUT) score have been reported to be useful predictors of morbidity and mortality for various oncological surgeries (12-16). These parameters are universally correlated with clinical outcome with high accuracy among various diseases, and have the advantage of being calculated easily without any invasive methods. Therefore, we considered whether these indicators could be applied to evaluate which elderly patients are high-risk PD patients.
Given the importance of assessing patient-specific risk factors for PD-related complications or mortality in the elderly, we conducted a retrospective study to investigate the impact of immunonutritional and physical parameters among patients undergoing PD. Specifically, we hypothesized that these parameters are useful to establish effective and safe strategies to prepare elderly patients for major abdominal surgery.
Patients and Methods
Patients. A retrospective cohort study was performed with 92 consecutive patients who underwent PD at a single institute (Kyushu Cancer Center, Fukuoka, Japan) from 2014 to 2018. Data from all of these patients were collected from medical charts and analyzed. The ages of the patients ranged from 42-87 years, and the mean and median ages were 68.3 years and 70 years, respectively. All patients basically sustained ability to perform certain activities of daily living, thus were categorized in Performance Status 0 or 1 by Eastern Cooperative Oncology Group scale. The co-morbidity scaled using Charlson's index (17) was 0/1 in 70 patients (76%) and ≥2 in 22 patients (24%). The patients were subcategorized into patients aged 79 years and younger (Group A, n=79), and 80 years and older (Group B, n=13). The mean/median ages of Group A and B were 65.9/68 and 82.5/82 years, respectively. The sex distribution was 57 (62%) male and 35 (38%) female. The disease was pancreatic tumor in 53 patients (58%), bile duct tumor in 29 (31%), and duodenal tumor in 10 (11%). The Child method was applied for the reconstruction after PD. An internal or external stent was placed during the pancreatojejunostomy, and the abdominal cavity was routinely drained using closed suction tubes, in all patients. Six patients (6%) underwent combined portal vein resection, 1 (1%) underwent hepatic artery resection and reconstruction, and 1 (1%) underwent partial hepatectomy. Postoperative pancreatic fistula and delayed gastric emptying were defined according to the International Study Group of Pancreatic Fistula (18, 19), and the grade of complication was defined by the Clavien-Dindo classification (20).
Immunological and nutritional assessment. Blood samples had been obtained within 1 week before surgery. Hematological parameters were determined immediately after blood sample collection, using an automatic blood analyzer. Parameters included white blood cell count, lymphocyte count, neutrophil count, monocyte count and platelet count (12, 21). The modified GPS was calculated from the serum albumin and serum C-reactive protein levels (22, 23). The PNI was calculated from the serum albumin level and the total lymphocyte count (24). Preoperative CONUT scores were calculated based on serum albumin concentrations, peripheral lymphocyte counts, and total cholesterol concentrations, as described elsewhere (16).
Body composition and physical functional assessment. Body composition and physical function assessments were performed by board certified physical therapists before surgery. The complete data was available and could be analyzed in 37 of the 92 patients, included 28 in Group A (79 years and younger) and 9 in Group B (80 years and older) patients. Body composition was determined using bioelectrical impedance analysis (BIA), with a TANITA MC-780A model body analysis monitor (Tanita Inc., Tokyo, Japan). Skeletal muscle mass (SMM) was calculated with the following equation: SMM (kg)=0.566 × fat free mass (25). Skeletal muscle mass index (SMI) was calculated as skeletal muscle mass (kg)/height squared (26). Cut-off thresholds for SMI were set at 7.0 kg/m2 and 5.7 kg/m2 for men and women, respectively (27). Handgrip strength (HGS) was measured to assess muscle strength, using a standard hand dynamometer. The cut-off thresholds for HGS were 18 kg for women and 26 kg for men (27). Walking speed was assessed via the 10 m walking test, and physical function was assessed using the 6-minute walking distance test (6MWD). The cut-off threshold for gait speed was 0.8 m/sec (27).
Statistical analysis. All statistical analyses were performed using JMP 14 software (SAS; Cary, NC, USA). All variables were expressed as the mean±standard deviation. Continuous variables were compared using the Mann–Whitney U-test. Categorical data were compared using the χ2 test. Logistic regression analysis was performed to identify the independent predictors of complications. Survival outcomes were determined using the Kaplan–Meier method and compared by the log-rank test. A multivariate Cox proportional model was used to analyze independent prognostic factors in both overall survival and recurrence-free survival. A p-value of <0.05 was regarded as statistically significant.
Results
Preoperative immunonutritional condition. Preoperative immunonutritional status was compared between Group A (79 years and younger) and Group B (80 years and older). We used an established index that has been reported to be associated with short- and long-term prognosis after various surgeries (16, 22-24). As shown in Table I, preoperative serum albumin was significantly lower in Group B. PNI and CONUT scores, which include albumin in their calculations, were significantly lower in Group B. The NLR was significantly higher in Group B, but there was no statistically significant difference between the two groups in the platelet-lymphocyte ratio and the lymphocyte-monocyte ratio. These data indicated a disturbed immunonutritional condition in the older patients.
Preoperative body composition and physical functional assessment. Body composition was assessed using BIA in 37 patients. The volumes and percentages of skeletal muscle were not different between the two groups, and thus SMI, the height-corrected skeletal muscle volume, was also similar between the groups (Table II). The body fat percentage tended to be higher in Group B (Table II, p=0.083). In contrast to body composition, a notable decrease of physical function was observed in the elderly group. HGS, gait speed, and 6MWD were significantly limited in the elderly patients (Table II). These results revealed that while the decrease of skeletal muscle volume was insignificant, the skeletal muscle function was dramatically impaired in the elderly compared to the non-elderly patients.
Comparison of immunonutritional parameters.
Comparison of body composition and functional assessment.
Comparison of operative parameters and postoperative complications between elderly and non-elderly.
Risk factors associated with Clavien-Dindo Grade ≥3 complications after pancreaticoduodenectomy.
Comparison of short-term outcome of surgery. The intra-operative parameters were compared between Groups A and B. The mean operative times and mean intraoperative blood losses were comparable in Groups A and B, and their transfusion requirements did not differ (Table III). All patients underwent potentially curative surgery. Then, the short-term postoperative outcomes of the surgery were compared. The rates of morbidities such as surgical site infection, pancreatic fistula, and delayed gastric emptying were comparable in Groups A and B. The incidence of postoperative delirium was significantly higher in the elderly group (p=0.0002). There was one postoperative death in Group A, for an overall mortality rate of 1.0%.
Risk factors associated with morbidity after pancreaticoduodenectomy. Since the morbidity rates were comparable between the elderly and non-elderly patients who underwent PD, we investigated risk factors associated with postoperative complications of Clavien-Dindo Grade ≥3 in all 92 patients. In multivariate analysis, soft pancreatic texture, high PNI (>46), and male sex were independent risk factors associated with postoperative complications (Table IV). High CONUT score and preoperative cholangitis tended be associated with postoperative complications (p<0.07).
Long-term prognosis. The overall survival rates and recurrence-free survival rates were similar between Group A and Group B (data not shown). We also investigated the survival rates in patients with pancreatic cancer (n=53), which revealed the same results. The multivariate analysis identified 3 factors that were prognostic for reduced overall survival (higher modified GPS index, higher CONUT score, and operative time >360 min) and one factor that was prognostic for reduced recurrence-free survival (postoperative complication, Clavien-Dindo Grade ≥3).
Discussion
We showed that immunonutritional index and physical function were significantly impaired in the elderly patients who underwent PD, although the rates of postoperative complications were comparable between the elderly and non-elderly patients. These results indicated that the elderly are likely to have developed frailty and sarcopenia. Thus, a prehabilitation program including nutritional support and physiologic intervention is strongly warranted for the elderly before they undergo PD, to prevent postoperative complications.
The safety of PD for the elderly is still controversial among previous reports, which means it is unclear whether the indication of surgery can be determined by chronological age. Pancreatectomy is still the only possible curative treatment for pancreatic and periampullary cancers, and thus it is necessary to re-evaluate the indication of PD in the elderly, considering the recent progress of perioperative management. Our data showed that the age category of 80 years and older was not associated with mortality and morbidity after PD; instead, morbidity was associated with impaired immunonutritional index (low PNI). The PNI was significantly lower in the patients 80 years and older. These results indicated that elderly patients should not be excluded on the basis their calendar ages. However, they also indicated that the decision to perform PD can be made on the basis of the physical and immunonutritional reserve, without regard to age.
Our data revealed that PNI and CONUT were significantly lower in the elderly who were to undergo PD, although these patients were predicted to be able to tolerate surgery. In addition to the nutritional index, NLR was also significantly higher in the elderly who were to undergo PD. A number of cohorts and meta-analyses supported the relationships between the immunonutritional index evaluated by laboratory data and long-term prognosis as well as postoperative complications in gastrointestinal and hepatopancreatobiliary (HPB) cancers (28, 29). The relevant biological mechanism relating the index and surgical outcomes has not been fully elucidated. However, it should be recognized that not only tumor-specific pathological factors, but also host-related nutritional status and systematic inflammatory response are responsible for the prognosis of patients with cancer. This study, to our knowledge, is the first to describe the relationship between aging and immunonutritional index in patients with HPB cancer.
It is well known that preoperative physical status, sarcopenia, comorbidities, and frailty are associated with surgical morbidity and mortality risk. Sarcopenia and frailty conceptually result from age-related decline in physiological reserve, and lead to disabilities, falls, fractures, and death (30). Our data showed that body composition and skeletal muscle amounts were comparable between the elderly and non-elderly subjects. These results may reflect that the patients were carefully selected to undergo PD, and the physical and nutritional prehabilitation protocols were introduced for the patients with sarcopenia and frailty to improve their physical condition in our series. Our data revealed that the physical condition, as evaluated by the grip strength and walking ability, was significantly impaired in the elderly, although body composition was comparable between the age groups. These results are worthy of attention because frailty caused by impaired physical activity has shown the most promise for identifying older adults at high risk of postoperative complications and nursing home discharge (31-33). We herein emphasize that physical activity, in addition to comorbidity, laboratory data and body composition, should be included in the preoperative geriatric assessment to decide on the operative indication for PD.
We acknowledge the limitations of this non-randomized study, but retrospective analysis is ethically and practically the only way to assess the eligibility of elderly patients for PD, given the evidence that surgery is the only curative option for pancreatic and periampullary cancer (34). Several recent studies reported that prehabilitation can reduce overall and pulmonary morbidity following major abdominal surgery (35, 36). We recommend further studies employing the prehabilitation protocol for elderly patients who undergo PD, to assess the efficacy of workup and to establish the optimal protocols for specific operative procedures or specific patients.
In conclusion, the present study justified PD for the elderly with acceptable morbidity and prognosis. However, immunonutritional status and physical function were significantly impaired among that group. Thus, appropriate case selection and active nutritional support are required in the elderly.
Acknowledgements
The Authors thank Ms. Y. Urasawa and Ms. K. Toyokawa for technical assistance, Ms. E. Shindo for body composition and physical functional assessment. This study was supported in part by the Uehara Memorial Foundation, Tokyo, Japan.
Footnotes
Authors' Contributions
Study conception and design: K Sugimachi; Acquisition of data: K Sugimachi, T Iguchi, Y Mano, M Ohta, M Ikebe; Analysis and interpretation of data: K Sugimachi, T Iguchi, Y Mano, T Nishijima; Drafting of manuscript: K Sugimachi, T Nishijima; Critical revision: M Morita, Y Toh.
Conflicts of Interest
K. Sugimachi and the co-authors have no conflicts of interest to declare regarding this study.
- Received September 20, 2019.
- Revision received September 30, 2019.
- Accepted October 1, 2019.
- Copyright© 2019, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.