To compare surgical and oncological outcomes after pancreaticoduodenectomy (PD) in patients ≥ 75 years of age with two younger cohorts of patients.

The prospectively maintained Institutional database of pancreatic resection was queried for patients aged ≥ 75 years (late elderly, LE) submitted to PD for any disease from January 2010 to June 2015. We compared clinical, demographic and pathological features and survival outcomes of LE patients with 2 exact matched cohorts of younger patients [≥ 40 to 64 years of age (adults, A) and ≥ 65 to 74 years of age (young elderly, YE)] submitted to PD, according to selected variables.

The final LE population, as well as the control groups, were made of 96 subjects. Up to 71% of patients was operated on for a periampullary malignancy and pancreatic cancer (PDAC) accounted for 79% of them. Intraoperative data (estimated blood loss and duration of surgery) did not differ among the groups. The overall complication rate was 65.6%, 61.5% and 58.3% for LE, YE and A patients, respectively, P = NS). Reoperation and cardiovascular complications were significantly more frequent in LE than in YE and A groups (P = 0.003 and P = 0.019, respectively). When considering either all malignancies and PDAC only, the three groups did not differ in survival. Considering all benign diseases, the estimated mean survival was 58 and 78 mo for ≥ and < 75 years of age (YE + A groups), respectively (P = 0.012).

Age is not a contraindication for PD. A careful selection of LE patients allows to obtain good surgical and oncological results.

Aging is a natural process and the number of elderly is rapidly growing in western countries. It has been estimated that Americans elderly are the fastest growing age group and that they will become more than a fifth of the whole population by 2030[1]. Hence, age-related chronic and neoplastic disease will be diagnosed always more frequently[2]. Pancreatic surgeons face daily with elderly patients. More than 60% of patients diagnosed with pancreatic cancer (PDAC) is ≥ 65 years[3,4]. In addition, PDAC is one of those cancers that will experience a 55% increase in diagnosis by 2030[5].

Pancreaticoduodenectomy (PD) is still burdened by high rates of morbidity and mortality, ranging from 35%-51% and 1%-6%, respectively[6-8].

Giving the unpredictable impact of aging on morbidity and mortality and considering the potentially higher risks of developing age-related complications[9,10], the decision to submit elderly to a PD can be challenging.

Two aspects should always be considered before submitting the elderly to a PD, first, for benign diseases, the post-operative recovery and the long-term quality of life; second, for malignancies, the fact that the elderly are less likely to receive proper adjuvant chemotherapy and this could make surgery less effective[11].

The aim of this study was to analyze the outcomes of PD on a population of elderly (≥ 75 years of age) and to compare them with 2 matched cohorts of younger patients (≥ 65 and < 75 of age and ≥ 40 and < 65 of age), to determine whether differences on surgical and oncological outcomes can be found in the index population.

The elderly population is increasing worldwide. In parallel, the rate of chronic and neoplastic diseases is rising steeply. PDAC, for whom aging is an independent risk factor[1], is one of those malignant neoplasms that will be diagnosed more frequently in the next decades[5]. In addition, the application of extensive cross-sectional imaging has led to an incredibly high prevalence of cystic pancreatic lesions[14], even in elderly. Thus, PD is always more frequently performed on elderly patients. In developed countries, the accepted definition of elderly population is of subjects ≥ 65 years of age. This cut-off considers social, cultural and working aspects. However, the biological status cannot be defined by a number and the decision to submit to surgery elderly patients must be weighed over the risks considering multiple factors (comorbidities, performance status, patient choice, home assistance, life expectancy).

Intraoperative data did not show any statistical difference in terms of duration of surgery, EBL and extent of lymphadenectomy. Based on this, we can claim that the surgical technique was not different in LE patients report an estimated mean survival of 70 mo. Patients ≥ 75 and < 75 years of age showed an estimated mean survival of 58 and 78 mo, respectively (P = 0.012). Senescence (n = 1) and cardiovascular problems (n = 3) were responsible for the 4 deaths reported within the LE group. On the contrary, no deaths were reported in the control groups after PD for benign conditions.

Considering overall post-operative complications, our data do not confirm previous findings of an overall higher incidence of complications in the elderly population[3,5-7,15-19]. In fact, the overall morbidity rate was homogeneous among the 3 groups. Interestingly, analyzing each complication considered, even the most fearsome such as PPH, CR-POPF, septic shock, there was not a statistically higher incidence in the LE group. However, the reoperation rate, of whom PPH, CR-POPF and septic shock are usually cause of, was statistically higher in the LE group (P = 0.003). Several considerations could explain this finding. First, the surgeon could unconsciously and paradoxically tend to be more aggressive with an elderly patient in the treatment of severe complications requiring surgery, hypothesizing functional reserves insufficient to allow for a conservative management of these complications. Second, in case of post-operative complications, pre-existent cardiopulmonary and vascular comorbidities typical of LE could precipitate the clinical conditions leading to reoperation. Third, the advanced age is associated with a higher risk of bacteremia[20], a condition that increases independently the risk of septic complications after PD[21]. When considering medical complications, the higher rate of cardiovascular problems registered in LE patients let us speculate that probably a dedicated perioperative fluid intervention with a negative fluid balance would prevent the onset of such disorders in this subset of patients.

Data regarding oncological outcomes after pancreatic surgery for malignancies in the elderly are controversial so far. Some Authors report that the elderly population shows a worse outcome[15,22] however other research groups did not find any significant survival difference[4,23]. This heterogeneity of results is probably due to selection biases.

In the present study, pooling all malignant tumors, the survival analysis revealed that LE patients had a trend towards a worse outcome compared with the other 2 groups, even if this was not statistically significant. However, the statistical significance was reached when LE patients were compared with the YE + A patients (P = 0.036), demonstrating that probably a larger sample size would have reach the statistical significance.

Stratifying the survival analysis for PDAC, LE patients, again, showed a trend towards a worse survival, however this data was not statistically significant, even comparing with the YE + A group. In our opinion, the most surprising data is the good overall survival reported for LE patients and the homogeneous distribution either of the different malignant diagnoses and the stage of the disease for PDAC among the 3 groups gives strength to this finding.

This finding is not surprising, since several factors can worsen the oncological outcomes in the elderly, such as a physiologic shorter life expectancy, pre-existent comorbidities that reduce functional reserves, worse post-discharge home care and rehabilitation services and a more difficult access to adjuvant therapy. In particular, a retrospective analysis by Sehgal et al[24] demonstrated that elderly patients receive less frequently adjuvant treatments after pancreatic surgery than younger patients, even if they obtain the same oncological benefits.

A separate mention should be made for survival data when PD was carried out for benign lesions. LE patients had a worse survival outcome if compared with < 75 years of age (57 mo vs 78 mo, respectively, P = 0.012). However, the small sample size and the fact the most of deaths were due to cardiovascular reasons, does not allow us to speculate that surgery could have somehow worsen or make evident pre-existent comorbidities leading to death. Nevertheless, it should always keep in mind that long-term postoperative consequences of PD, such as exocrine pancreatic insufficiency with malabsorption and loss of weight, new onset or worsening of pre-existent diabetes and alteration in diet, are managed with more difficulty by LE patients than by younger patients. In addition, LE patients often live alone because of widowhood and the adherence and medication management could be insufficient.

Our results strengthen the idea that advanced age does not represent per sé a contraindication for PD and, in general, for pancreatic surgery. However, considering the higher rates of reoperation and cardiovascular complications found in LE patients we can claim that this subgroup of patients shows a latent frailty that requires a global careful attention, to prevent the development of life-threatening complications. On the other side, present survival data for LE patients demonstrate that submitting to a PD a LE patient for a periampullary malignant tumor is at least “worthwhile”.

This study has biases that must be considered, such as the retrospective nature, the lack of data about the number of LE patients who did not receive surgery due to comorbidities or to patient refusal, the lack of data regarding the performance status of LE patients, the high number of missing data about adjuvant therapy. In addition, our population of LE patients was well-selected (as demonstrated by the median ASA score of 2) and we probably submitted to PD the most fit LE. At the same time, the strength of this research is the case-matching with 2 different cohorts, that allowed us to give weight to advanced age as independent variable.

In conclusion, the advanced age is not an independent contraindication to PD. A fit for surgery group of LE patients can be submitted to PD safely and fruitfully. Even more caution should be used when surgery is indicated for benign conditions, where life expectancy is higher.