As discussed above, the goal of CRS, whether it be pCRS or iCRS is to achieve NGR. This outcome is achieved in a variable proportion of patients dependent on presenting disease burden and surgeon skill and comes at a significant cost in terms of morbidity and mortality. Many single institution studies have been published exploring this issue in the “elderly patient” with a variety of definitions of elderly and a variety of outcomes – pointing to the dearth of prospective data in the area (Table 4.2).

Because of the retrospective nature of these works and the variety in endpoints studied, it is difficult to summarize outcomes for anything other than postoperative death which seems to increase markedly with age. If we classify elderly as > 65, prevalence of postoperative death is 0–10.3 % (with 0 % in the series with almost 50 % of patients undergoing iCRS) [12, 20, 23]. If we change the cutpoint for elderly to >70, the range is 0–9 % [2, 6, 8, 32, 35, 38]. Finally, if we set the cutpoint at age >80, postoperative death is reported in 0–13 % of series [10, 24, 25].

In order to make the more informed decisions regarding how to care of elderly patients with ovarian cancer and provide them with the most accurate counseling, we need good estimates of the following 4 outcomes: (1) postoperative death, (2) postoperative morbidity, (3) non-traditional discharge, and (4) deviations from or elimination of planned chemotherapy.

Among the possible complications following CRS, postoperative death is the most significant. Case series report prevalence of 0–13 % with significant variability in populations. Thrall et al. published a SEER-Medicare evaluation which sought to identify the prevalence of 30-day postoperative mortality among patients undergoing CRS and develop a predictive nomogram for this complication. This study contained only women >65 years and included those undergoing both pCRS and iCRS. They found an overall 30-day postoperative mortality rate of 8.22 %. However, when patients who were admitted and underwent surgery emergently were evaluated separately, the rate of 5.56 % among scheduled and 20.12 % among emergent cases. Age was highly associated with postoperative death in both these groups. When evaluated as a continuous variable among patients admitted for scheduled surgery, each year over age 65 was associated with a 7.5 % increase in the risk of 30-day mortality (95 % CI 1.06–1.10). Among patients admitted and operated on emergently, each year over age 65 was associated with a 2.8 % increase in the risk of 30-day mortality (95 % CI 1.01–1.05).

Among those patients admitted for elective surgery who had a history of NACT, the 30-day mortality was 70 % lower than those who underwent pCRS (1.82 vs. 6.13 %).

When evaluating patients admitted for elective surgery with no history of NACT, the risk of 30-day mortality could be divided into low (<5 %), intermediate (5–10 %), and high (>10 %) risk based on age, stage, and comorbidity score. The highest risk group included patients 75 years of age or older, with stage IV disease or stage III disease plus a comorbidity score of 1 or more. This group comprised 25.7 % of the patients and had a 30-day postoperative mortality of 12.7 % which accounted for 50 % of the deaths in the entire cohort [33].

This risk score has not been validated in prospective trial but is valuable information for evaluating patients and making treatment decisions based on the best information available.

Postoperative complications are important to avoid, not only for the obvious reason of eliminating patient discomfort and risk but also because certain complications are associated in a reduction in overall survival. This is not to say that patients die of their complications, which would be postoperative death, but the complication compromises the ability of the patient to survive their cancer. National Surgical Quality Improvement Program (NSQIP) data found that the most important determinant of decreased survival following surgery for patients with solid tumors with occurrence of a major complication [16]. These data exist in ovary cancer as well where the occurrence of 2 or more complications increased the risk of death from cancer (HR 1.31 95 % CI 1.15–1.49) [40]. Wright et al., evaluating the Nationwide Inpatient Sample registry of patients admitted for ovarian cancer surgery, found that perioperative complications increased with age. This study could not separate patients who underwent pCRS and those who underwent iCRS and so they are presented together. Among the 28,651 patients included, those patients < 50 years of age had a 17.1 % complication rate as compared to 29.7 % in patients aged 70–79 and 31.5 % in patients > 80. These trends held regardless of categorization as a medical, surgical, or infectious complication. Not surprisingly, rates of complications also increased by age and by the number of radical procedures performed at the time of surgery. Patients over 70 who required 2 or more radical procedures (bowel resections, diaphragm resections, etc) at the time of surgery had complications in 33 % of cases. One radical procedure in the age range resulted in 25 % with major complications. Given the commonality of radical procedures during pCRS to achieve NGR, these are useful estimates in counseling patients and determining the best sequence of therapy [39] (Table 4.3).