European Urology

1. Introduction

Serious illness, surgery, and postoperative stress response place increased demands on the body's nutritional requirements. These conditions promote a catabolic state that puts the patient at risk of malnutrition. Malnutrition, in turn, leads to higher morbidity, longer hospitalisation, and a higher incidence of hospital-acquired infections [1] and [2]. Because biochemical signs of malnutrition appear rapidly after major surgery, many intensive care units, especially in Europe, have adopted total parenteral nutrition (TPN) as part of postoperative care following major surgery—including cystectomies [3]—in up to 42% of patients despite adequate gastrointestinal function and even though benefits have only been demonstrated in malnourished patients or in well-nourished patients enduring a postoperative period of semistarvation exceeding 7–10 d [1], [4], and [5]. Following radical cystectomy (RC), TPN is also used to prevent hypoproteinemia caused by the extensive wound area. This is especially important, because hypoproteinemia leads to oedema of the bowel wall and thus aggravates the postoperative ileus, a common postoperative complication occurring in up to 25.8% of cystectomy patients [6]. Whether recovery following RC can be improved by postoperative TPN, however, has never been clearly shown [7], [8], [9], and [10]. In the present prospective, randomised trial, we evaluated the clinical impact of TPN versus oral alimentation alone following extended pelvic lymph node dissection (ePLND), cystectomy, and ileal diversion.

2. Patients and methods

2.1. Study population

Between September 2008 and March 2011, 169 consecutive bladder cancer patients scheduled for ePLND, cystectomy, and ileal diversion were assessed for eligibility to enter the study. Exclusion criteria were previous pelvic lymph node dissection (PLND), chronic inflammatory bowel disease, previous radiation therapy, prior bowel surgery, severe hepatic (ascites, portal hypertension, encephalopathy), or cardiac dysfunction (New York Heart Association class greater than III) as well as an inability to give fully informed consent. One hundred fifty-seven patients were included in the study and prospectively randomly allocated by a computer-based program into two groups: group A (n = 74) received TPN, and group B (n = 83) received oral nutrition alone (Fig. 1). Patients with a body mass index (BMI) <18.5, serum albumin <30 g/l, serum prealbumin <0.2 g/l, serum total protein <60 g/l, and/or involuntary weight loss >10% in the preceding 6 mo were defined as malnourished. The study was approved by the local ethics committee. All patients gave their informed consent.

Fig. 1 Consort diagram showing the numbers of patients who were randomly assigned, received intended treatment, and were analysed.

3. Results

Preoperative patient characteristics were similar, and the types of urinary diversion (UD) performed were evenly distributed. The median time for surgery and blood loss did not differ between the two groups (Table 1). Postoperative complications occurred in 51 patients (69%) in group A versus 41 patients (49%) in group B (p = 0.013; Fig. 2). There were significantly more infectious complications in group A than in group B (32% vs 11%; p = 0.001). There was no difference in the rate of complications other than infectious complications (p = 0.789; Table 2). The majority of complications (62%) were minor (Clavien-Dindo grade 1 or 2). There were no intraoperative deaths, but early lethal complications (grade 5) occurred in one patient from each group (1%; Fig. 2).

Fig. 2 Early postoperative complications patients experienced by highest grade according to the five-grade Clavien-Dindo classification system [10].

In all patients, serum albumin, serum prealbumin, and serum total protein dropped significantly after surgery and gradually increased until postoperative day 12 (Fig. 3A–3C). Serum prealbumin and serum total protein were significantly lower in group B than in group A on postoperative day 7 (Fig. 3B–3C). These differences in biochemical parameters had vanished by postoperative day 12. CRP rose significantly after surgery in both groups, showing significantly lower peaks on postoperative day 3 in group A than in group B (p < 0.001; Fig. 3D).

Fig. 3 Postoperative course of biochemical nutritional and inflammatory parameters: (A) serum albumin, (B) serum prealbumin, (C) serum total protein, and (D) C-reactive protein.CRP = C-reactive protein.

Except for fewer episodes of vomiting on postoperative day 3 in group B, nutrition type had no effect on gastrointestinal function (Fig. 4A–4D). Median time to resumption of full diet (9 d vs 9 d; p = 0.978) as well as the length of hospital stay (16 d vs 15.5 d; p = 0.365) did not differ between the two groups. The costs attributed to TPN were €614 per patient.

Fig. 4 Early postoperative follow-up. Parameters were assessed on postoperative days 1, 3, 7, and 12: (A) vomiting, (B) bowel movements, (C) flatulence, and (D) passage of stool.

A post hoc analysis performed on the subgroup of patients classified preoperatively as malnourished (n = 34) showed no significant difference in overall (82% vs 52%; p = 0.096) or infectious (18% vs 13%; p = 0.692) complication rates between group A malnourished patients (n = 11) versus group B malnourished patients (n = 23). Except for a significantly lower serum prealbumin level on postoperative day 7 in group B (median 0.19 g/l vs 0.14 g/l; p = 0.011), no differences in nutritional or inflammatory biochemical variables were observed between the two malnourished subgroups.

4. Discussion

RC combined with ePLND and UD is an extensive surgical procedure causing a high rate of postoperative complications [6]. The complications are related to many factors, including surgical trauma, intestinal oedema, and blood loss. The present randomised study was initiated to determine whether TPN can reduce the incidence of postoperative complications. Surprisingly, our findings show that postoperative TPN was associated with an increased rate of infectious complications as well as with higher costs. This finding is in line with the findings of a large meta-analysis on parenteral nutrition [8] and of another meta-analysis analysing the outcome of early enteral versus parenteral nutrition [9]. Possible reasons for the adverse effect of TPN are first, that parenteral nutrition impairs the immune system by causing dysfunction of B and T lymphocytes, macrophages, and neutrophils [13]. Second, TPN raises blood glucose levels, making insulin treatment necessary, which in turn appears to have a negative impact on the immune system and increases susceptibility to infection [9] and [13]. Third, TPN leads to intestinal mucosal atrophy with consequent increased intestinal permeability favouring bacterial translocation; in contrast, enteral nutrition prevents adverse structural and functional alterations of the gut barrier, improves intestinal blood flow, and increases the systemic and local immune response [14], [15], and [16]. Finally, TPN increases the risk of catheter-related bloodstream infections [9] and [13].

In contrast to the assumption of Maffezzini et al. that early postoperative provision of artificial nutrients, whether in the form of TPN or enteral nutrition, has beneficial effects [17], our data show that TPN is associated with a higher rate of infectious complications. Some patients, however, experience prolonged postoperative ileus and do not tolerate oral intake in the first postoperative days. In these critically ill patients, parenteral nutrition should be considered if (semi)starvation lasts for >7–10 d [4] and [5] so as to prevent the serious adverse effects of long-term protein-energy malnutrition.

Elevated CRP is the most important indicator of acute-phase reaction. It is generally acknowledged that early postoperative CRP plasma concentrations are an indicator of the degree of trauma inflicted upon the organism by surgery, with a peak on postoperative day 3 [18]. This finding is in line with our findings. The significantly lower CRP values on postoperative day 3 in our patients receiving TPN contrast with the higher rate of infection in this group. Nevertheless, >90% of the infectious complications occurred later than postoperative day 3. A possible explanation is that TPN may have contributed—at least to some extent—to the lower CRP levels by negatively influencing the function of T lymphocytes and macrophages, both of which secrete interleukin-6, which in turn stimulates CRP production. However, lower CRP values in patients receiving postoperative TPN versus patients receiving postoperative enteral nutrition alone was not found in other studies [16] and [19].

Serum prealbumin and serum total protein were restored earlier in group A patients. However, this result had no impact on clinical outcome. This finding is in contrast to the results of Maffezzini et al. [7], who did not find any effect of early postoperative parenteral nutrition on postoperative protein depletion. However, they measured serum albumin and serum total protein only on postoperative days 3 and 5, whereas we could show that postoperative normalisation of the serum proteins takes ≥7 d. This delay is most probably the result of transcapillary escape of serum proteins into the interstitial fluid during the first postoperative days, a phenomenon that largely depends on the surgical trauma and its consequent systemic inflammatory response syndrome [20].

Like Maffezzini et al. [7], we found no significant difference in time to recovery of bowel function in our patients. The 13% incidence of postoperative ileus—defined as postoperative vomiting associated with abdominal distension requiring cessation of oral intake and/or permanent opening of the gastrostomy tube or the intolerance of oral intake of our patients until postoperative day 5—is in line with the findings of others [6] and [21]. Although delayed postoperative return of bowel function is acknowledged to be the most frequently encountered complication after RC, high variability exists in the incidence of postoperative ileus, most probably because of the different definitions used for this condition [6], [7], and [22]. We therefore analysed not only the incidence of postoperative ileus but also other objective, measurable parameters such as bowel movements, flatulence, and passage of stool, although these parameters have their own observer and time-dependent weaknesses.

Our overall 59% complication rate also agrees with the results of another major centre using a standardised prospective reporting methodology [6]. The majority of complications (62%), however, were minor (grade 1 or 2; Fig. 2). At the time the present prospective study was performed, our anaesthetists routinely administered high volumes of parenteral crystalloids perioperatively. Doing so is known to significantly increase postoperative complications as well as postoperative body weight resulting from, for example, intestinal oedema [23] and might have negatively influenced perioperative blood loss. The extent to which this high-volume perioperative fluid management may have negatively influenced our results remains unknown; we recently abandoned it in favour of increased use of vasoactive agents to compensate for the anaesthesia/analgesia-induced vasodilation.

A potential limitation is that we only used a standard TPN solution containing polyamino acids, glucose, and electrolytes. Other parenteral nutritional supplementations are available (eg, with lipids), although several reports have shown that lipids adversely affect immune status, complications, and clinical outcome [24]. In addition, more inflammatory and immunologic markers could have been used. It should be noted, however, that the main focus of the study was the patients’ clinical outcome. Another limitation is that the study had to be stopped early because of the potential harm of TPN, which may have influenced results to some extent.

There is no evidence that withholding parenteral nutrition for at least 7 d is harmful in patients who are not severely malnourished [8]. Still, precise diagnosis of malnutrition is difficult, because there is no standardised and widely accepted definition. This major methodological shortcoming contributes to the heterogeneity of findings among studies [25], [26], [27], and [28]. Using our definition for malnutrition, the prevalence of preoperative malnutrition in our patients was 22%. In contrast to the findings of some authors [1], [2], and [4] but in line with Bozzetti et al. [29] and Mohler et al. [30], our post hoc subgroup analysis could not demonstrate any clinical benefit of TPN for malnourished patients with regard to preventing postoperative complications. Still, our subgroup analysis is underpowered and does not allow for any definitive conclusions.

5. Conclusions

The findings of the present prospective, randomised study show that postoperative TPN following ePLND, cystectomy, and ileal diversion has considerable disadvantages in terms of a higher rate of complications—mainly infectious—when compared to postoperative oral alimentation alone. We therefore cannot recommend TPN as part of standard postoperative management following ePLND, cystectomy, and ileal diversion.

Author contributions: Urs E. Studer had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Studer, Burkhard, Huwyler, Thalmann, Roth.

Acquisition of data: Roth, Huwyler, Zehnder, Birkhaeuser.

Analysis and interpretation of data: Roth, Studer.

Drafting of the manuscript: Roth.

Critical revision of the manuscript for important intellectual content: Studer, Birkhaeuser, Zehnder.

Statistical analysis: Roth, Studer (external analysis).

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Studer, Burkhard, Thalmann.

Other (specify): None.

Financial disclosures: Urs E. Studer certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: None.

Trial registration: ClinicalTrials.gov; ISRCTN79535150.