Effect of Postoperative Restrictive Fluid Therapy in the Recovery of Patients with Abdominal Vascular Surgery

Article Outline

• Abstract
• Methods
  • Patients
  • Fluid regimens
  • Statistical analyses
• Results
• Discussion
• References
• Copyright

Abstract 

Objective

To compare the outcome of the postoperative administration of a restricted or standard intravenous fluid regimen in patients who underwent elective abdominal vascular surgery. The primary end point was postoperative hospital stay.

Design

Prospective observer-blinded, randomised controlled trial.

Material and methods

Patients were considered eligible if they underwent transperitoneal aorto-iliac approach with infrarenal graft repair. During the operation and intensive care unit stay, fluids were prescribed by the anaesthetists who were unaware of the details of the study. In the vascular surgical ward, patients in the standard group (SG) received 2500ml of fluids per day, whereas patients in the restricted group (RG) received 1500ml of fluids per day. All the patients were evaluated on an intention-to-treat basis.

Results

Forty patients were randomised to the RG (n=20) or SG (n=20). No significant differences were observed in the recovery of gastrointestinal function. However, the postoperative hospital stay was shorter in the RG (8 days) than in the SG (12 days) (p=0.003).

Conclusions

The use of a restrictive postoperative fluid protocol significantly reduces the duration of hospital stay in patients who have undergone major elective abdominal vascular surgery.

Keywords: Fluid therapy, Perioperative fluid balance, Vascular complications

Although the use of intravenous fluids is an important part of perioperative care in patients undergoing vascular surgery, few studies have evaluated this issue. A large variability in fluid regimen has been noted throughout the surgical specialties, but it has been a common practice to administer relatively large amounts of fluids, regardless of blood loss, body weight or anaesthetic technique.¹ In addition, the task of prescribing fluid and electrolytes is often left to the most junior member of the team, leading to significant variability in prescribing fluids which can result in adverse events, particularly in patients at the extremes of their life span.² Fluid minimisation regimens in patients undergoing elective major surgery, particularly colorectal surgery, reduce perioperative complications and improve postoperative outcome.³, ⁴, ⁵ This study aimed to compare the outcome of the postoperative administration of a restricted (1500ml of fluid per day) or standard (2500ml of fluid per day) intravenous fluid regimen in patients undergoing elective open abdominal vascular surgery, in the belief that a restrictive fluid therapy would be associated with better postoperative recovery.

Back to Article Outline

Methods 

The study protocol was approved by the institutional review board and local research ethics committee. A prospective, observer-blinded, randomised controlled trial was carried out between January and December 2007 in our university teaching hospital. Informed, written consent was obtained, and patients undergoing elective open abdominal vascular surgery were randomised before the operation to either the restricted intravenous fluid (1500ml per day) or standard care (2500ml per day) group during their postoperative stay in the surgical ward. A computer-generated random number pattern, in blocks of four, was used for the allocation sequence. The primary end point was the duration of postoperative hospital stay.

Patients 

Adult patients due to undergo elective abdominal vascular surgery were eligible if they were scheduled for the transperitoneal aorto-iliac approach, through a standard midline laparotomy incision, with infrarenal graft repair. We excluded patients who were pregnant or had mental disorders, severe physical disability, impaired renal function, congestive cardiac failure, hepatic disease, cancer, inflammatory bowel disease or were receiving drugs that affect gastrointestinal motility.

All patients received bowel preparation (a phosphate enema) the night before and were allowed free fluids until 12h before the surgery. A homogeneous anaesthetic protocol was followed in all cases because they were all managed by a single anaesthetist. Normothermia was maintained throughout the surgery, and nasogastric tubes and intra-abdominal drains were used. The cell saver was not employed. All patients received antibiotic prophylaxis (cefazoline 2g e.v. in the operation theatre, and 1g – every 8h for the initial postoperative 24h). At the end of the procedure, all patients were shifted to the intensive care unit (ICU) for at least 24h and then returned to the specialist beds in the vascular surgery unit for the rest of the postoperative period. As determined by randomisation, the patients were then treated with a standard or restrictive fluid protocol. Paracetamol (1g, 4 times daily e.v.) was administered concurrently with tramadol (50mg intravenously) for breakthrough pain, if necessary. Epidural analgesia was only maintained during the ICU stay. The consumption of analgesics and blood-component levels as well as the water–electrolyte balance were noted.

Operative factors that were identified include the nature of surgery (occlusive or aneurysmal disease), duration of surgery, clamp time and blood loss. The total fluid intake – intravenous fluid (including blood and blood products) and oral fluids – from midnight on the day of the operation (day 0) to the discharge date was recorded. The total fluid output (including urine, enteral and surgical drain losses) during the same period was recorded. Outcome measures included the length of postoperative stay (including the ICU and vascular ward), fluid balance (daily and cumulative) and bowel motility (occurrence of first bowel sounds, first flatus and resumption of first intake of liquid and solid food intake and passage of first stool). A biochemical analysis was performed preoperatively (baseline) and on the 3rd and 7th day after the operation. We also recorded all adverse events and re-admissions during the first 30 postoperative days.

Fluid regimens 

During the operation and postoperative ICU stay, fluids were prescribed by the anaesthetists who were blinded to the details of the study and the randomisation. These regimens were similar in both groups and managed by individualised haemodynamic monitoring. Generally, 500ml of lactated Ringer's solution was preloaded, prior to epidural analgesia, and saline (0.9%) replaced the loss to third-space intra-operatively. External losses were replaced. Approximately 1000–1500ml of saline (0.9%) replaced blood loss up to 500ml, and hydroxyethyl starch (HAES; 6%) replaced additional blood loss. Blood-component therapy began when estimated blood loss approximated 1500ml or the haematocrit decreased significantly. In the ICU, the fluid administration usually consisted of 3000ml per day (1000ml of saline (0.9%) and 2000ml of dextrose (5%) with potassium supplementation if required).

In the vascular surgical ward, patients in the standard group (SG) received 2500ml of fluids per day (1000ml of dextrose (5%) and 1500ml of saline (0.9%) with 40mmol of potassium), whereas patients in the restricted group (RG) received 1500ml of fluids per day (saline (0.9%) with 40mmol of potassium). In both regimens, blood transfusions were administered if needed. The goal was to achieve a haematocrit greater than 30%. Antibiotics and analgesia were dissolved in saline and included in the saline volume. All patients received chest physiotherapy and commenced active mobilisation from the 2nd postoperative day. Oral fluids were encouraged after the 3rd day following the operation, and administration of antiemetic medication was recorded.

Clinical decisions about discontinuation of intravenous fluids, resumption of diet and discharge were made by the treating surgical team and not by the investigators. The investigators were blinded to the treatment of each patient and did not review the patient in the ward. Discharge delayed by social problems was recorded as such. To be considered fit for discharge, patients had to be apyrexic, fully mobile, passing flatus or faeces and on oral medication.

Statistical analyses 

All randomised patients were analysed on an intention-to-treat basis. Before the trial, we estimated that a sample of 17 patients in each group was required to detect a reduction in hospital stay of 20%, with 80% power at a significance level of p=0.05. Mean (standard deviation, SD) values were calculated for continuous variables and categorical data were expressed as absolute numbers with percentages. We used the Mann–Whitney U test, Friedman test, chi-square test and Fisher's exact test to determine the significance of differences between the groups. All statistical analyses were performed using SPSS software (version 14.0; SPSS, Chicago, IL, USA).

Back to Article Outline

Results 

During the study period, 43 eligible patients were identified of these, 40 gave their consent and were randomised to the RG (n=20) or a standard postoperative fluid regimen group (SG, n=20) (Table 1). The reasons for non-randomisation were anaesthetic cancellations (two cases) and patient refusal (one case). No patient in this series violated the randomisation sequence. Twenty-six patients with atherosclerotic occlusive disease were treated by the aortobifemoral bypass graft (RG: 14 cases, SG: 12 cases), while 14 patients with infrarenal aortic aneurysms were treated by resection and graft interposition (RG: 6 cases, SG: 8 cases). No significant surgical differences were found between both the groups. The mean operating time was 198.2±52min in the SG compared with 196.5±37min in the RG (p=0.640), with an ICU stay duration of 1.90±1.7 days and 1.75±0.6 days, respectively (p=0.529) (Table 2).

Table 1. Baseline characteristics

	Standard group (n=20)	Restrictive group (n=20)	p
Age (years, CI 95%)	61.95 (56.7–67.2)	65.50 (62.1–68.9)	0.398
Sex (male/female)	17/3	20/0	0.231
Risk factors
-Diabetes	6 (30%)	6 (30%)	1.00
-Hypertension	11 (55.0%)	13 (65%)	0.519
-Hypercholesterolaemia	8 (40%)	10 (50%)	0.525
-Cardiac disease	5 (25%)	9 (45%)	0.185
-COPD	7 (35%)	4 (20%)	0.288
-Smoker	14 (70%)	14 (70%)	1.00
ASA classification
-ASA-1	1 (5%)	0	0.551
-ASA-2	8 (40%)	9 (45%)
-ASA-3	11 (55%)	10 (50%)
-ASA-4	0	1 (5%)
Diagnosis
-Occlusive	15 (75%)	12 (60%)	0.311
-AAA	5 (25%)	8 (40%)

COPD: chronic obstructive pulmonary disease; ASA: American Association of Anaesthesiology; AAA: abdominal aortic aneurysm.

Table 2. Surgical factors and hospital stay

ICU: intensive care unit.

The mean balance for the RG and SG was also similar on the day of surgery and on the subsequent days in the ICU (6600.5±2227.9ml in the SG compared with 5970.4±2531.4ml in the RG, p=0.40). Blood loss did not differ in either group; in fact, the incidence of blood transfusions was similar in both groups (405.0±367.7ml in the SG vs. 336.1±433.3ml in the RG, p=0.60).

However, the mean postoperative balance in the surgical ward was lesser in the RG (16.8±2026.4ml) than in the SG (1980.7±2327.4ml) (p=0.007) (Table 3). None of the patients received artificial nutritional support or blood transfusions during their stay in the vascular ward. Patients in both groups received similar amount of antibiotics or analgesia. There were no differences in the time to passage of first flatus (p=0.883), bowel sounds (p=0.841), resumption of liquid intake (p=0.192), resumption of solid food intake (p=0.277) and passage of first stool (p=0.192) between the study groups (Table 4). Likewise, no significant postoperative complications were observed, although a re-intervention (thrombectomy), one pulmonary oedema and two wound infections were recorded in the SG, and more electrolyte alterations occurred in the SG than in the RG. However, this study was not powered to these end points.

Table 3. Cumulative water balance

	Standard group		Restrictive group		p
	Mean	CI 95%	Mean	CI 95%
Perioperative balance (ml) (OR–ICU stay)	6600.5	5557.8–7643.2	5970.4	4785.6–7155.1	0.409
-Water input	11311.3	9435.4–13187.3	11314.6	9281–13348.1	0.998
-Water output	4710.8	3703.3–5718.3	5344.2	4068.7–6619.6	0.420
Postoperative balance (ml) (surgical ward)	1980.7	891.4–3070.0	16.8	−931.5–965.2	0.007
-Water input	10773.2	8780.5–12765.9	5797.5	4581.5–7013.4	0.000
-Water output	8792.5	6634.7–10950.3	5780.6	4556.0–7005.2	0.015
Total balance	8581.2	6877.8–10284.6	6049.7	4709.7–7389.5	0.019

OR: operating room; ICU: intensive care unit.

Table 4. Time in relation to gastrointestinal function (days)

No patient developed renal failure (oliguria) or suffered any adverse events as a result of biochemical changes. In fact, the serum urea, creatinine, osmolality, albumin and haemoglobin levels did not differ in the postoperative period between the groups (Table 5).

Table 5. Biochemical changes

	Standard group		Restrictive group		p
	Mean	CI 95%	Mean	CI 95%
Blood urea (mmoll−1)
-Basal	46.90	39.35–54.45	48.25	41.28–55.22	0.678
-Third postoperative day	34.75	25.52–43.98	38.90	31.17–46.63	0.301
-Seventh postoperative day	29.05	18.56–39.54	31.95	26.47–37.43	0.134
Serum creatinine (mmoll−1)
-Basal	0.94	0.85–1.02	1.05	0.94–1.16	0.096
-Third postoperative day	0.89	0.75–1.02	1.00	0.87–1.13	0.211
-Seventh postoperative day	0.81	0.69–0.93	0.87	0.76–0.97	0.429
Serum osmolality (mOsmkg−1)
-Basal	289.8	287.30–292.30	290.65	289.06–292.24	0.552
-Third postoperative day	281.15	277.17–285.13	285.65	282.55–288.75	0.070
-Seventh postoperative day	285.60	282.62–288.58	286.05	282.40–289.70	0.843
Serum albumin (gl−1)
-Basal	4.025	3.875–4.171	3.965	3.750–4.180	0.678
-Third postoperative day	2.740	2.527–2.953	2.941	2.703–3.179	0.220
-Seventh postoperative day	3.055	2.792–3.318	3.088	2.848–3.329	0.862
Haemoglobin (gl−1)
-Basal	14.44	13.62–15.25	14.95	14.15–15.75	0.277
-Third postoperative day	11.63	10.61–12.64	11.26	10.68–11.83	0.968
-Seventh postoperative day	11.59	10.75–12.43	11.61	10.99–12.23	0.841

The time at which patients were considered fit for discharge was reduced in the RG than in the SG (8.40±1.39 days vs. 12.40±7.95 days, p=0.003). No patient required re-admission to the hospital within 30 days of surgery, although a death secondary to cardiac problems was recorded in the SG during the follow-up (at home, 18th day).

Back to Article Outline

Discussion 

This study shows that patients receiving significantly more fluids in the postoperative period develop a longer hospital stay, despite having no significant differences in surgical factors.

Current clinical practice has been largely influenced by the theory of a decrease in extracellular volume after surgery, due to internal redistribution of fluids (loss to third space), and the replacement of these losses by additional fluid infusion has been advocated. This has led to the liberal use of intravenous fluids postoperatively, and it has not been uncommon to see large amounts of fluid being administered in elective surgical procedures far exceeding the losses.¹, ⁶, ⁷ This is especially the case in major aortic surgery. In fact, in our study, there was a mean over-hydration of 6l of fluid during operation and ICU stay. When considering perioperative fluid management, it is important to reach a balance between administering too little fluid, with consequent hypovolaemia and organ dysfunction, and too much fluid, with resulting oedema and organ dysfunction.⁵ The wide range of fluids given to the protocol groups emphasises the fact that giving a single volume load is not appropriate when applied to individuals. Fluids must be individually indicated in relation to appropriately monitored haemodynamic changes.⁸

Administration of excess fluid may cause several problems after surgery¹, ⁹, ¹⁰, ¹¹: it may potentially increase the incidence of postoperative cardiac morbidity, respiratory failure or pneumonia by fluid accumulation in the lungs and gastrointestinal motility may be inhibited, thus prolonging postoperative ileus. In addition, excess fluid may decrease tissue oxygenation with adverse implications for wound healing and, finally, coagulation may be enhanced with crystalloids which may predispose patients to postoperative thrombosis. Although our study was not powered to analyse the postoperative complications between groups, it was observed that a fewer number of patients developed side effects in the RG than in the SG. Probably due to this fact, the time at which patients were considered fit for discharge was significantly reduced in the RG (8 days) than in the SG (12 days). On the other hand, we did not observe differences in relation to the recovery of gastrointestinal function between the groups.

One of the major concerns with postoperative fluid restriction is the potentially unrecognised (or subclinical) hypovolaemia resulting in organ dysfunction.⁵ Although patients in the RG received less water than those in the SG, none of the patients became oliguric or had a concentration of creatinine in blood that was above the upper limit of normal range. In fact, the biochemical markers did not differ in the postoperative period between groups. However, the administration of 1500ml of fluid in the RG was arbitrary, since the appropriate volume to be administered is yet to be evaluated, and such a protocol would also allow for the eventuality of bleeding.

The lack of side effect of our postoperative restriction regimen on clinical outcomes may be explained by the perioperative overload in the operating room and during the ICU stay. The gross fluid gained into the extracellular space maintained the haemodynamic stability in the surgical ward, passing water and sodium from the interstitial to the intravascular space. This could account for a shorter hospital stay. In contrast, in the SG, the postoperative over-hydration combined with the neuroendocrine response to the surgery – which helps to retain water and sodium and to increase the microvascular permeability – made these patients vulnerable to extravascular fluid accumulation, with consequent oedema, more complications and a longer hospital stay.

Fluid management should be accorded the same status as drug prescription.⁷ We suggest an approach that would facilitate goal-directed fluid supplementation during surgery, with fluid restriction for uncomplicated patients starting early in the postoperative period.⁸ Simply following a protocol that sets limits on fluid administration (as in the present study) is an alternative approach, which can be further adapted by the use of more invasive monitoring in high-risk patients.⁸, ¹² In fact, current opinion suggests that maintaining fluid balance such that weight remains stable may constitute the best practice in postoperative fluid management.⁵

This study may have some limitations. First, the relatively small sample of homogeneous patients in one institution may limit the generalisability of the results. Second, the daily weight was not monitored, an important variable to monitor the postoperative fluid balance of each patient. Finally, some confounding factors (type of surgery, anaesthetic protocol, ICU stay, etc.) may present difficulties in the interpretation of the results.

In conclusion, our study shows that the use of a restrictive postoperative fluid protocol significantly reduces the duration of hospital stay in patients undergoing major elective abdominal vascular surgery.

CONTRIBUTORS (providing patients): Carpintero LA, Gutiérrez-Alonso V, Carrera-Díaz S, De Marino P, Del Blanco I, Del Rio L, Martín-Pedrosa M, Ibáñez MA and San Norberto E.

Conflict of interest/funding: none.

Back to Article Outline

References 

1. Holte K, Sharrock NE, Kehlet H. Pathophysiology and clinical implications of perioperative fluid excess. Br J Anaesth. 2002;89:622–632
   • View In Article
   • MEDLINE
   • CrossRef
2. Lobo DN, Dube MG, Neal KR, Allison SP, Rowlands BJ. Perioperative fluid and electrolyte management: a survey of consultant surgeons in the UK. Ann R Coll Surg Engl. 2002;84:156–160
   • View In Article
   • MEDLINE
3. Lobo DN, Bostock KA, Neal KR, Perkins AC, Rowlands BJ, Allison SP. Effect of salt and water balance on recovery of gastrointestinal function after elective colonic resection: a randomised controlled trial. Lancet. 2002;359:1812–1818
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (193 KB)
   • CrossRef
4. Brandstrup B, Tonnesen H, Beier-Holgersen R, Hjortso E, Ording H, Lindorff-Larsen K, et al. Effect of intravenous fluid restriction on postoperative complications: comparison of two postoperative fluid regimens: a randomized assessor-blinded multicenter trial. Ann Surg. 2003;238:641–648
   • View In Article
   • MEDLINE
   • CrossRef
5. Mackay G, Fearon K, McConnachie A, Serpell , Molloy RG, O'Dwyer PJ. Randomized clinical trial of the effect of postoperative intravenous fluid restriction on recovery after elective colorectal surgery. Br J Surg. 2006;93:1469–1474
   • View In Article
   • MEDLINE
   • CrossRef
6. Brandstrup B, Svensen C, Engquist A. Hemorrhage and operation cause a contraction of the extracellular space needing replacement – evidence and implications? A systematic review. Surgery. 2006;139:419–432
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (202 KB)
   • CrossRef
7. Lobo DN, Macafee AAL, Allison SP. How perioperative fluid balance influences postoperative outcomes. Best Pract Res Clin Anaesthesiol. 2006;20:439–455
   • View In Article
   • MEDLINE
8. Sartain J. Pathophysiology and clinical implications of perioperative fluid excess (correspondence). Br J Anaesth. 2003;90:395–396
   • View In Article
   • MEDLINE
   • CrossRef
9. Lowell JA, Schifferdecker C, Driscoll DF, Benotti PN, Bistrian BR. Postoperative fluid overload: not a benign problem. Crit Care Med. 1990;18:728–733
   • View In Article
   • MEDLINE
   • CrossRef
10. Nisanevich V, Felsenstein I, Almogy G, Weissman C, Einav S, Matot I. Effect of intraoperative fluid management on outcome after intraabdominal surgery. Anesthesiology. 2005;103:25–32
    • View In Article
    • MEDLINE
    • CrossRef
11. McArdle GT, Price G, Lewis A, Hood JM, McKinley A, Blair PH, et al. Positive fluid balance is associated with complications after elective open infrarenal abdominal aortic aneurysm repair. Eur J Vasc Endovasc Surg 2007;34:522–7.
    • View In Article
12. Bundgaard-Nielsen M, Holte K, Secher NH, Kehlet H. Monitoring of perioperative fluid administration by individualized goal-directed therapy. Acta Anaesthesiol Scand. 2007;51:331–340
    • View In Article
    • MEDLINE
    • CrossRef