Anesthetics and neuraxial anesthesia commonly result in vasodilation/hypotension. Norepinephrine counteracts this effect and thus allows for decreased intraoperative hydration. The authors investigated whether this approach could result in reduced postoperative complication rate.
In this single-center, double-blind, randomized, superiority trial, 166 patients undergoing radical cystectomy and urinary diversion were equally allocated to receive 1 ml·kg−1·h−1 of balanced Ringer’s solution until the end of cystectomy and then 3 ml·kg−1·h−1 until the end of surgery combined with preemptive norepinephrine infusion at an initial rate of 2 µg·kg−1·h−1 (low-volume group; n = 83) or 6 ml·kg−1·h−1 of balanced Ringer’s solution throughout surgery (control group; n = 83). Primary outcome was the in-hospital complication rate. Secondary outcomes were hospitalization time, and 90-day mortality.
In-hospital complications occurred in 43 of 83 patients (52%) in the low-volume group and in 61 of 83 (73%) in the control group (relative risk, 0.70; 95% CI, 0.55–0.88; P = 0.006). The rates of gastrointestinal and cardiac complications were lower in the low-volume group than in the control group (5 [6%] vs. 31 [37%]; relative risk, 0.16; 95% CI, 0.07–0.39; P < 0.0001 and 17 [20%] vs. 39 [48%], relative risk, 0.43; 95% CI, 0.26–0.60; P = 0.0003, respectively). The median hospitalization time was 15 days [range, 11, 27d] in the low-volume group and 17 days [11, 95d] in the control group (P = 0.02). The 90-day mortality was 0% in the low-volume group and 4.8% in the control group (P = 0.12).
A restrictive-deferred hydration combined with preemptive norepinephrine infusion during radical cystectomy and urinary diversion significantly reduced the postoperative complication rate and hospitalization time.
Norepinephrine counteracts the vasodilating effects of anesthesia, thus allowing blood pressure to be maintained with less fluid administration
In a randomized trial, the investigators compared major outcomes in patients given conventional amounts of fluid with fluid sparing induced by norepinephrine administration
The incidence of major complications was significantly reduced, with a relative risk of 0.7 (95% CI, 0.55–0.88)
The duration of hospitalization was also significantly reduced from a median of 17 to 15 days
IT is well established that perioperative fluid management influences postoperative complication rates in major gastrointestinal surgery.1–4 However, the optimal intraoperative fluid regimen is still controversial in terms of how much fluid to infuse, choice of fluids (crystalloids or colloids), use of vasopressors, or a goal-directed hemodynamic therapy (GDT).1–4
Restrictive intraoperative hydration has shown ambiguous results, some showing a reduced postoperative morbidity,1,3 whereas others found no benefit in terms of reduction of complications.5–7 However, the impact of intraoperative restrictive deferred crystalloid hydration combined with the administration of preemptive continuous norepinephrine infusion on the postoperative complication rate has not yet been investigated in patients undergoing major gastrointestinal surgery. Norepinephrine, which has a strong α- and only a mild β-adrenergic effect, may counteract anesthesia-induced vasodilation and consequent hypotension and ensure sufficient organ perfusion.8 This could be a valuable alternative to IV fluid boluses, as liberal hydration is known to cause interstitial edema and increased postoperative complications.1,3
To test this hypothesis in a major surgery model we chose open radical cystectomy combined with pelvic lymph node dissection and urinary diversion, because it is an extensive and time-consuming but standardized surgical intervention in a high case-load tertiary center. In addition, despite improvements in surgical technique and perioperative care, radical cystectomy is associated with early postoperative complication rates of 26–64%9–13 and a 90-day mortality rate of 2–7%.12–14
The goal of this study was to compare the impact of intraoperative restrictive deferred hydration combined with preemptive infusion of norepinephrine versus that of a more liberal crystalloid hydration (including fluid preloading during induction and “generous” intraoperative hydration) on in-hospital complication rates, hospitalization time, 90-day postoperative complications, and mortality in patients undergoing pelvic lymph node dissection, radical cystectomy, and urinary diversion.
Material and Methods
Trial Design and Participants
This prospective, single-center, double-blind (patient, data assessor, and surgeon), parallel-group, randomized, controlled superiority trial was approved by the local ethics committee (Kantonale Ethikkommission Berne, KEKBE154/08) and registered at ClinicalTrials.gov (NCT01276665). After providing written informed consent, 190 consecutive patients were assessed for eligibility between November 2009 and September 2012 (fig. 1).
Patients with American Society of Anesthesiologists physical status II or III undergoing pelvic lymph node dissection and open radical cystectomy with either an ileal conduit or an ileal orthotopic bladder substitution for urinary diversion were included. Exclusion criteria were known coagulopathies, significant hepatic dysfunction (prothrombin ratio <50%), significant renal dysfunction (estimated glomerular filtration rate <60 ml/min), congestive heart failure (New York Heart Association scores ≥3), and contraindications for epidural analgesia.
Randomization and Masking
The randomization sequence was created by computer-generated permuted block randomization with 1:1 allocation. The allocation sequence was concealed in opaque sealed envelopes that were sequentially numbered. The randomization sequence was kept concealed until after written consent had been obtained. Patients were allocated to the groups by assigning them the sequentially numbered envelope with the lowest number. Stratification was done for the type of urinary diversion (ileal conduit vs. ileal orthotopic bladder substitution). Patients, urologists, and data assessors were blinded to group assignment. In order to blind the surgeons, crystalloid bags and perfusion pumps were placed behind an opaque panel during surgery and were not visible to the urologists. Data were recorded prospectively in a standardized case report form. Assessors of the postoperative data had no access to the anesthesiologic patients’ data as these were kept sealed.
Preoperative Care, Surgery, Anesthesia, and Monitoring
No antegrade bowel preparation was used. Patients fasted till midnight and could drink clear drinks till 2 h before surgery. All patients were premedicated with midazolam (7.5 mg) or lorazepam (1 mg) 30 min before induction of anesthesia. Surgery was standardized and performed as previously described with the patient in a 30° head-down position in the presence of one senior urologist (Drs. Burkhard, Thalmann, and Studer).15–17
Standard monitoring included continuous electrocardiographic data, heart rate, nasopharyngeal core temperature, pulse oxymetry, invasive mean arterial pressure (MAP) with a radial artery catheter and central venous pressure with a venous catheter inserted in the right jugular vein. For study purposes, an esophageal Doppler probe (Deltex Medical Ltd., Chichester, United Kingdom) was inserted immediately after induction of anesthesia and placed to obtain the best possible Doppler velocity signal from the descending aorta. Stroke volume, cardiac output, and corrected flow time were recorded. An epidural catheter was placed at the T9/T10 level: an 18-gauge epidural needle was inserted by a paramedian approach and the epidural space was identified with the loss-of-resistance technique. After a test dose of 1.5 ml lidocaine 2% with 0.005 mg/ml epinephrine to rule out subarachnoidal or intravascular placement, a 0.25% bupivacaine infusion at a rate of 8 ml/h was administrated until the end of the pelvic lymph node dissection and then stopped until closure of the abdominal wall. Anesthesia was induced with propofol (2 mg/kg), fentanyl (2 µg/kg), rocuronium (0.6 mg/kg) and maintained with isoflurane at an age-corrected minimum alveolar concentration of 0.6. Ventilation with an inspired oxygen fraction of 60% was mechanically controlled to maintain ParterialCO2 between 35 and 40 mmHg, with a positive end-expiratory pressure of 5 mmHg and tidal volume of 8 ml/kg. Normothermia was maintained with a convective air warming system (Bair Hugger; 3M-Switzerland, Rüschlikon, Switzerland) and a Hotline® fluid warmer (Smith Medical International Ltd., Ashford, Kent, United Kingdom).
Intraoperative Fluid Therapy
Restrictive Deferred Fluid Group with Preemptive Norepinephrine Administration (“Low-volume Group”).
After induction of anesthesia, a preemptive norepinephrine infusion was started at a rate of 2 µg·kg−1·h−1 until the end of surgery and a balanced Ringer’s solution (Ringerfundin®; B. Braun Medical AG, Sempach, Switzerland) was infused at a rate of 1 ml·kg−1·h−1 until the bladder was removed, followed by 3 ml·kg−1·h−1 of balanced Ringer’s solution until the end of surgery (deferred hydration). If hypotension was observed (MAP <60 mmHg), norepinephrine infusion rate was titrated to maximum 8 µg·kg−1·h−1 after an initial bolus of 10 µg. If hypotension persisted, a bolus of 250 ml of balanced Ringer’s solution was given.
A preload bolus of 6 ml/kg of balanced Ringer’s solution was administrated during induction of anesthesia. After that, balanced Ringer’s solution was infused at a constant rate of 6 ml·kg−1·h−1 until the end of surgery. If hypotension was observed (MAP <60 mmHg) a bolus of 250 ml balanced Ringer’s solution was given and in case of persistent hypotension this procedure was repeated to a maximum of 10 boluses.
In both groups, blood loss of greater than 500 ml was substituted with an equal amount of balanced Ringer’s solution. Packed erythrocytes units were transfused if hemoglobin values were less than 8 g/dl (<10 g/dl in patients with coronary artery disease). Fresh frozen plasma transfusion was administrated in the presence of continuous excessive microvascular bleeding (relative indication based on the senior surgeons’ observations) or if prothrombin time was greater than 1.5 times normal.18 Colloid solution (Voluven balanced®; Fresenius Kabi AG, Stans, Switzerland) was only infused as a rescue medication if a MAP less than 60 mmHg persisted after the abovementioned correction with balanced Ringer’s solution, and in case of severe metabolic acidosis (base excess <−5, pH <7.25) attributable to severe hypovolemia.
Intra- and Postoperative Biomarkers
Serum lactate and central venous oxygen saturation (Scvo2) were assessed after induction of anesthesia, after cystectomy, after urinary diversion, at the end of surgery, and on postoperative day (POD) 1. Hemoglobin, hematocrit, C-reactive protein, procalcitonin, albumin, and creatinine levels were assessed preoperatively, on PODs 1, 2, and 5. The cardiac biomarkers high-sensitive troponin T, brain natriuretic peptide (BNP), serum and urine osmolality, were assessed preoperatively, on PODs 1 and 2. At hospital discharge, hemoglobin, hematocrit, creatinine, and BNP serum levels were assessed.
Postoperative Patient Management
The epidural analgesia was reactivated during closure of the abdominal wall with a mixture composed of bupivacaine 0.1%, fentanyl 2 µg/ml, and epinephrine 2 µg/ml using a CADD-Legacy ambulatory infusion pump (model 6300; Deltec Inc., St. Paul, MN). The initial infusion rate was 8 ml/h, with a maximum infusion rate till 15 ml/h, and with additional bolus volumes of 5 ml (lockout time: 1 h). After surgery patients were admitted to the intermediate care unit. IV paracetamol and metamizol (both 1 g every 6 h) were given postoperatively. The epidural catheter was removed on POD 5.
Postoperative hydration was identical in both groups and consisted primarily of 1,000 ml of balanced Ringer’s solution and 500 ml of glucose 5% per 24 h until resumption of normal food intake.19 If the MAP were less than 60 mmHg, first, a bolus of 500 ml of balanced Ringer’s solution was administrated, second, by persistent MAP less than 60 mmHg, norepinephrine was infused up to a rate of 200 µg/h. Packed erythrocytes units were transfused according to the American Society of Anesthesiologists guidelines,20 and fresh frozen plasma transfusion was given if the prothrombin time was greater than 1.5 times the normal value. The amounts of packed erythrocytes units, fresh frozen plasma units, and fluids given postoperatively were recorded.
Postoperative patients were allowed to drink clear fluids immediately. A peroral liquid diet was started on POD 1 as well as active mobilization. To enhance recovery of bowel function, the use of chewing gum was encouraged21 and subcutaneous application of neostigmine 0.05 mg was started on POD 2. Flatus and stool passage were recorded. Body weight was measured daily. Perioperative antibiotic therapy consisted of obramycin and metronidazole for 2 days and amoxicillin/clavulanic acid until removal of all stents and catheters. Low–molecular-weight heparin was started on the evening before surgery and maintained throughout hospitalization.
The primary endpoint was the in-hospital complication rate assessed according to a modified postoperative morbidity survey and Clavien–Dindo classification for radical cystectomy3,22 (appendix). Secondary endpoints were hospitalization time, the 90-day postoperative complication rate according to the Clavien–Dindo classification,10,23 and 90-day mortality. Start of the hospitalization time was defined as the moment the patient was admitted into the hospital (in all cases the day before surgery). Discharge criteria were: removal of all drains and catheters, and the ability to handle the urostoma bag or empty the ileal orthotopic neobladder spontaneously and free of residual urine.
All outcome measures were registered by assessors blinded to the intraoperative fluid regimen.
On the basis of the assumption that the application of intraoperative restrictive deferred fluid regimen with preemptive norepinephrine infusion would reduce the complication rate from 3811–13 to 20%, a sample size of 83 patients per group was calculated for each group with a type I error of 0.05 and a power of 80%. Data were analyzed on a modified intention-to-treat basis (one patient was excluded after randomization because the initially planned surgery was cancelled). Data are expressed with medians with ranges for continuous variables or frequencies for categorical ones. Categorical data were compared with the Fisher exact or the chi-square test and continuous data with the Mann–Whitney U test as appropriate. Relative risks (RRs) and 95% CIs were also calculated. The primary outcome was analyzed using Fisher exact test with RR and 95% CI. The significance level was set at 0.05.
Multiple logistic regression analyses using a full model were applied to identify independent risk factors for postoperative complications and reported as adjusted odds ratios (ORs) with 95% CIs. Confounders considered were groups (low-volume vs. control group), type of urinary diversion (ileal conduit vs. ileal orthotopic blabber substitution), age (categorized into: <65 yr, 65–74 yr, 75–84 yr, ≥85 yr), neoadjuvant chemotherapy, preoperative anemia, American Society of Anesthesiologists physical status score (II vs. III), Charlson Comorbidity Index age adjusted, Glasgow Prognostic Score, body mass index, and preoperative serum BNP values. Interaction terms were not included because of insufficiently large sample size. Confounders were considered as significant if their P values were less than 0.10. The fit and predictive capability of the model was assessed using the Hosmer–Lemeshow goodness-of-fit test and receiver operating characteristic area under the curve. The statistical softwares used were SPSS 19.0 (SPSS Inc., Chicago, IL) and Statistical Analysis System software (version 9.3; SAS Institute, Cary, NC).
Of 190 consecutive patients, 167 fulfilled the eligibility criteria and were randomized. One of these patients was excluded because the intervention was aborted, leaving 166 patients included in the final analysis (fig. 1). Complete follow-up data were available for all 166 participants. Preoperative and patient characteristics were similar for the two groups (table 1).
The total median volume of balanced Ringer’s solution infused intraoperatively was 1,700 ml [range, 700, 4,000 ml] in the low-volume group versus 4,300 ml [2,800, 6,200 ml] in the control group; P < 0.0001 (table 2). The median dose of norepinephrine administered intraoperatively to the low-volume group was 3.6 µg·kg−1·h−1 [2.0–8.0 µg·kg−1·h−1]. Neither group was given colloid solution. Significantly lower median cardiac output values were observed in the low-volume group than in the control group during the periods of cystectomy (3.7 l/min [2.2–8.7 l/min] vs. 4.8 l/min [1.7–11.9 l/min]; P = 0.002) and urinary diversion (4.2 l/min [2.4–7.3 l/min] vs. 5.0 l/min [2.3–11.1 l/min]; P = 0.003) but not at the end of surgery (table 2). In both groups, cardiac output values at the end of surgery did not differ significantly from values at the beginning of surgery. On POD 1, a significantly greater increase in body weight was observed in the control group (2.0 kg [−2, +7 kg]) than in the low-volume group (0.0 kg [−3, +4 kg]; P < 0.0001; table 3).
In-hospital complications occurred in 43 of 83 patients (52%) in the low-volume group versus 61 of 83 patients (73%) in the control group (RR, 0.70; 95% CI, 0.55–0.88; P = 0.006). The absolute reduction in the number of patients experiencing complications was 22% (95% CI, 7–36%). The total number of complications was 77 in the low-volume group versus 161 in the control group (P < 0.0001; table 4). According to the Clavien–Dindo classification, the majority of complications were minor (Clavien–Dindo classification grade 1 or 2) in both groups: low-volume group (33 of 83 patients; 40%) and control group (42 of 83 patients; 51%; fig. 2).
Gastrointestinal complications occurred in 5 of 83 (6%) of the low-volume group versus 31 of 83 (37%) of the control group (RR, 0.16; 95% CI, 0.07–0.39; P < 0.0001). The most common complications observed were ileus and constipation, which occurred in 0 of 83 (0%) and 2 of 83 (2%) patients of the low-volume group versus 8 of 83 (10%) and 18 of 83 (22%) of the control group (P = 0.007 and P = 0.0006, respectively; table 4).
Cardiac events occurred in 17 of 83 patients (20%) of the low-volume group versus 40 of 83 patients (48%) in the control group (RR, 0.43; 95% CI, 0.26–0.69; P = 0.0003). The majority of cardiac events were minor complications consisting of a transient increase of BNP during the two first PODs: 11 of 83 (13%) patients in the low-volume group versus 28 of 83 (34%) patients in the control group (RR, 0.39; 95% CI, 0.21–0.74; P = 0.003). The incidences of the major complications such as acute myocardial infarction and congestive heart failure did not differ between the two groups, nor did the rates of renal, infectious, pulmonary, and thromboembolic complications (table 4).
Multiple logistic regression analyses identified group allocation (low-volume vs. control group: OR, 0.44 [95% CI, 0.21–0.92]; P = 0.029), body mass index (per increasing value: OR, 1.13 [95% CI, 1.01–1.25]; P = 0.029), BNP (per increasing value BNP: OR, 1.01 [95% CI, 1.00–1.03]; P = 0.045), preoperative anemia (no vs. yes: OR, 0.426 [95% CI, 0.18–1.02]; P = 0.056), and Glasgow Prognostic Score (per increasing value: 1.61 [95% CI, 0.93–2.88]; P = 0.089) as independent predictors of complications (Hosmer–Lemeshow test: P = 0.685 and receiver operating characteristic area under the curve: 0.78; table 5).
Length of Hospital Stay, 90-day Postoperative Complication Rate, and Mortality
The median length of hospital stay was 15 days [11, 27d] in the low-volume group and 17 days [10, 95d] in the control group (P = 0.01).
The total number of patients experiencing complications within the 90th POD was lower in the low-volume group (44 of 83; [53%]) than in the control group (64 of 83 [77%]; P = 0.0019). The majority of complications in both groups were classified as minor (grade 1 or 2; table 6).
The 30-day and 90-day overall mortality rates were 0 and 2.4%, respectively. No patient in the low-volume group died, but four patients (4.8%) in the control group did (P = 0.12). Two patients died because of rapid disease progress, one died of septic shock, and one of pneumonia. Three of these patients were older than 75 yr.
Intraoperative restrictive deferred hydration combined with preemptive norepinephrine infusion was associated with significantly reduced in-hospital and 90-day postoperative complication rates and a shortened hospitaliza tion time.
Consistent with other reports10,24 constipation and ileus were the most frequently observed gastrointestinal complications in our patients receiving more liberal crystalloid hydration, despite the use of the modified Clavien–Dindo classification for radical cystectomy,10 which has rarely been applied in other studies. In the low-volume group, gastrointestinal complications were significantly lower with a more rapid recovery of bowel function, which is of relevance because gastrointestinal dysfunction can lead to prolonged hospitalization.22,25 The postoperative body weight increase and the hypoalbuminemia observed in the control group reflect overhydration, which is known to delay recovery of bowel motility because of excess fluid in the intestinal wall.19 Excess of fluid leads to shedding of the glycocalyx followed by extravasation and edema.26
The number of cardiac events was high in both groups, but significantly lower in the low-volume group; however, it has to be stated that in both groups most cardiac events were a transient increase of serum BNP. Severe cardiac complications like arrhythmia, congestive heart failure, and acute myocardial infarction did not differ between the groups. Serum BNP release is directly proportional to ventricular volume expansion, and thus the transient BNP increases observed in the control group may reflect a perioperative volume overload and correlates with Berri et al.’s27 observations, where a postoperative serum BNP increase was associated with a positive fluid balance in patients undergoing pancreatectomy. In addition, a postoperative BNP increase has been associated with prolonged hospitalization time in patients undergoing major orthopedic surgery.28 The number of arrhythmias and myocardial infractions, complications that could be attributed to norepinephrine, were not different between the groups. Our results show that the use of norepinephrine combined with a low-volume fluid regimen as administered in our study was safe and not detrimental to cardiac function, as has often been postulated.
A frequently used argument against restrictive intraoperative hydration is the fear of increased infectious complications attributable to hypovolemia and the ensuing tissue hypoperfusion.2 In our low-volume group, the rate of infectious complications was not increased. On POD 1, the patients’ body weight and urine osmolality, factors that mirror hydration status, were comparable with the preoperative values, indicating a zero fluid balance rather than hypovolemia. An Scvo2 value less than 70% is considered an early marker for tissue hypoperfusion and is associated with increased postoperative complications.29,30 Scvo2 remained greater than 70% in our low-volume group, indicating an adequate balance between oxygen delivery and consumption. Serum lactate is another indicator of tissue oxygenation, and a serum lactate greater than 2 mm indicates acidosis. The low number of patients in both groups with postoperative serum lactate levels greater than 2 mm and the median serum lactate levels of less than 1.5 mm observed perioperatively are comparable with those levels in studies using hemodynamic optimization strategies (GDT).2,31 The nonsignificant difference in infectious complications further corroborates adequate tissue perfusion and the safety of the restrictive deferred hydration combined with preemptive norepinephrine infusion. The significantly lower 90-day complication rate in the low-volume group is particularly noteworthy in these patients who are at high risk for complications and documents the safety of the restrictive deferred hydration combined with preemptive norepinephrine infusion. The complication rate in the control group is in line with the complication rates reported by other high case-load institutions, despite the fact that in this study complications were graded according to the modified Clavien–Dindo classification.10,32 This is of importance because it indicates that the complication rate in the control group, although significantly higher than in the low-volume group, was not excessive when compared with rates in other centers of excellence.
Although serum creatinine values were increased in the low-volume group on POD 1 (most likely because of less serum dilution), renal function (serum creatinine values, estimated glomerular filtration rate, and urine osmolality) did not differ between the groups from POD 2 until discharge. Fluid restriction and the adjuvant use of norepinephrine did not result in more renal complications and a more liberal fluid administration did not appear to have a protective effect on renal function. On the basis of these findings, the often assumed beneficial effect of preoperative fluid boluses and liberal fluid administration to protect renal function seems questionable.33 It has been shown that intraoperative urinary output is dependent not only on fluid management but also on the intraoperative release of stress hormones, osmolality, and adiuretin secretion.34
Major concern is often voiced concerning the vasoconstrictive effect of norepinephrine. Our results document, however, that the preemptive norepinephrine infusion at an initial rate of 2 µg·kg−1·h−1 has no identifiable negative consequences. On the contrary, it counteracts the decreased sympathetic tone and vasodilatation induced by epidural analgesia, anesthetics, and analgesics and may be more physiological at compensating for a plegic vascular system than the liberal use of IV fluids. This is also substantiated by the overall lower complication rate in this group. In addition, although postoperative serum BNP and transient creatinine level increases were rated by us as complications, which makes difficult the comparison with other series that did not include these parameters, the incidence of overall complications, using similar Clavien–Dindo classification, in the low-volume group was still lower than that of other major urological centers.10,32 Our results call into question the current policy of liberal fluid administration to replace basic fluid volume depleted by, e.g., perspiration, exudation through the surgical wound, loss into the third space, and to correct hypotension due to vasodilation.1,3,6,35
Besides the preemptive and concomitant use of norepinephrine, our low-volume group protocol differed from protocols of precedent studies because of the deferred fluid administration: less fluid was administrated during pelvic lymph node dissection and cystectomy (the main period of bleeding); thereafter during construction of the urinary diversion more fluid was administered. This may explain why blood loss was significantly reduced in the low-volume group and consequently, less packed erythrocytes units was transfused perioperatively. This again could explain the enhanced recovery in the low-volume group.36
A possible limitation of this study is that the low-volume group is not compared with a GDT protocol, a mode of fluid management that has been shown to enhance bowel recovery and shorten hospitalization time after abdominal surgery.25 The GDT approach, which consists of a baseline fluid substitution with additional fluid boli to optimize cardiac preload, is generally associated with more fluid administration (including colloids), than administered in our control group. In addition, the routine or preemptive use of norepinephrine administration was not a first-line option in GDT protocols in contrast to our protocol.2,25,37,38 In a previous randomized trial comparing the effect of GDT (14 ml·kg−1·h−1) with that of standard intraoperative fluids administered at the discretion of the anesthesiologist (11 ml·kg−1·h−1) on bowel function after radical cystectomy, a significant reduction of ileus was noted in the GDT group (22 vs. 53%).38 By comparison, the rates of constipation and ileus in both our groups were far lower. This could be explained not only because of the intraoperative hydration, but also by the fact that in the study by Pillai et al.,38 not every patient had an epidural analgesia and it was instituted only for 48 h. In addition, our finding is in line with a recent study showing that GDT and a zero fluid balance fluid regimen have similar complication rates after colorectal surgery.6 Our results using intraoperative restrictive deferred hydration combined with preemptive norepinephrine infusion raise questions concerning the validity of GDT or liberal intraoperative fluid administration (as in Pillai et al.’s control group) during open radical cystectomy and urinary diversion.
A minor limitation of this study is the lack of continuous Scvo2 monitoring. We only measured Scvo2 at three time-points during surgery and thus, cannot exclude that Scvo2 values less than 70% occurred intraoperatively in both groups.
A strength of this study is that because all patients were managed with the same postoperative enhanced recovery protocol, the risk of possible confounding factors caused by individualized postoperative management is limited. And because no patients were lost to postoperative follow-up, it is unlikely that any complications were missed.
In conclusion, restrictive deferred fluid management combined with preemptive norepinephrine administration in patients undergoing standardized pelvic lymph node dissection, open radical cystectomy, and urinary diversion resulting in a postoperative zero fluid balance, lower in-hospital and 90-day postoperative complication rates, and reduced hospitalization time.
The authors thank Professor Juerg Huesler, Ph.D. (Institute of Mathematical Statistics and Actuarial Science, University of Bern, Bern, Switzerland), for help with the statistical analyses.
Support was provided by the Else Kröner Fresenius Foundation (2010_A80), Bad Homburg, Germany, and institutional research funds of the Department of Anesthesiology and Pain Therapy and of the Department of Urology, University Hospital Bern, Bern, Switzerland.
The authors declare no competing interests.
Appendix. Definitions of Postoperative Complications/Events
Ileus: no evidence of bowel function (no flatus and no passage of stool) with abdominal distension requiring cessation of oral intake and intravenous fluid support by POD 510
Constipation: no passage of stool without signs of ileus by POD 510
Gastric ulcer: diagnosis made by gastroscopy
Anastomotic bowel leak: considered as a complication if requiring surgery or prolonged drainage
Complications of Infections
Urinary tract infection: temperature >38°C in the last 24 h and leukocytosis and a prompted urinary analysis that showed bacterial counts >100,000 requiring antibiotics
Sepsis: bacterial infection and at least two of the following clinical signs: hypo- or hyperthermia, tachycardia, tachypnea, leukocytopenia or leukocytosis, positive blood culture3
Pneumonia: temperature >38°C and leukocytosis and clinical signs of pneumonia, requiring antibiotics
Wound infection: pus can be expressed or aspirated, requiring surgical intervention
Wound dehiscence: diagnosed clinically and requiring resuturing
Myocardial infarction: increase of the enzyme high-sensitive troponin T above the hospital laboratory’s myocardial infarction threshold (>0.05 µg/l), and either new Q wave changes, or persistent changes in ST-T segments
Arrhythmia: confirmed by 12-lead electrocardiography and requiring new medication or electroconversion
Congestive heart failure and pulmonary edema: shortness of breath, rales, jugular venous distension, peripheral edema, third heart sound, radiologic signs (cardiomegaly, interstitial or alveolar edema), brain natriuretic peptide value >500 pg/ml and diagnosis requiring diuretics
Transient brain natriuretic peptide increase: postoperative serum brain natriuretic peptide values between 100 and 500 pg/ml (considered as minor cardiac event)
Pulmonary embolism: evidenced by spiral computerized tomography scanning3
Renal dysfunction: transient increase of creatinine: creatinine >50% upper limit of normal value3
Renal failure: severe reduction in glomerular filtration rate (15–29 ml·min−1·1.73 m−2) at discharge
Urinary leakage: radiologically diagnosed, requiring stenting
Presence of a de novo focal deficit, confusion/delirium