3.3.4.3 COMPLICATIONS

SURGERY

• The Cochrane Colorectal Group (2012) have evaluated the literature on preoperative nutritional support in patients undergoing GI surgery.³³⁸ Three studies were included that compared preoperative PN with no nutrition support in these patients, most of whom were malnourished. The meta-analysis showed a significant reduction in major post-operative complications from 45% in the control group who received no nutrition support to 28% in the group receiving PN (RR 0.64; 95% CI 0.46–0.87), with low heterogeneity.
• Two RCTs published before Cochrane review but that were not included support this finding. An RCT by Bozzetti and colleagues (2000) that evaluated perioperative TPN in malnourished patients with GI cancer showed that 10 days’ preoperative TPN, continued postoperatively, significantly reduced the rate of complications by approximately one-third compared with no preoperative nutrition (37% vs 57%; p = 0.03); non-infectious complications accounted for most of this difference (12% vs 34%; p = 0.02).⁶¹ In addition an RCT by the Veterans Affairs TPN Study group involving 395 malnourished patients who required laparotomy or non-cardiac thoracotomy found that severely malnourished patients who received perioperative TPN had significantly fewer non-infectious complications than patients who did not receive TPN (5% vs 43%; p = 0.03).⁶⁰
• A meta-analysis of seven studies (n = 798) by Braunschwieg and colleagues (2001) to evaluate the benefits of PN versus standard care (conventional oral diet and IV dextrose) in patients undergoing surgery for cancer of the oesophagus or stomach found that in studies with high percentages of malnutrition, PN was associated with a trend toward a lower risk of infection ((RR 1.17 for standard care vs PN; 1.17, 95% CI 0.88–1.56).³³⁹
• Similarly, another meta-analysis of randomised studies in patients undergoing surgery for cancer of the oesophagus or stomach reported by Braunschweig et al. (2004) demonstrated a trend for reduced infections and complication rates with PN in studies of malnourished patients (RR 1.17; 95% CI 0.88–1,56 for standard care vs PN) and a reduction in post-operative complications.
• Heyland and colleagues (2001) carried out a meta-analysis of 27 randomised trials in surgical patients, to compare use of TPN versus standard care (usual oral diet plus IV dextrose).⁶⁴ TPN was associated with significantly fewer major complications (RR 0.81; 95% CI 0.65–1.01), particularly in malnourished patients.
• In the randomised study by Xian-li and colleagues (2004) that compared glutamine- supplemented TPN, standard TPN and no TPN in 64 patients with serious acute pancreatitis receiving traditionally therapy,³²⁴ the incidence of complications (acute respiratory distress syndrome, multi-system organ failure) was significantly higher in patients receiving no TPN (21%) than in those receiving standard TPN (11%; p < 0.01) or glutamine-supplemented TPN (4%; p < 0.01). Furthermore, length of stay was significantly shorter for patients receiving either standard TPN (28.6 days; p < 0.05) or glutamine-supplemented TPN (25.3 days; p < 0.01) compared with no TPN (39.1 days).
• Thoracic chyle leakage is a major complication of oesophagectomy. In a prospective study by Weijs and colleagues (2017), consecutive patients with chyle leakage (n = 371) were assigned to a low-fat diet (for leakage <500 mL/day), TPN (for leakage >1,000 mL/day), or a low-fat diet or TPN (for leakage 500–1,000 mL/day) depending on whether the chyle leakage was increasing or decreasing at diagnosis and the patient’s condition.³⁴⁰ Treatment could then be stepped up (e.g., TPN or surgery) based on outcomes with the initial treatment. A low-fat diet was the initial treatment for 61 patients (78%) and was successful in 40 (66%) after a median of 9 days. TPN was subsequently administered to 20 of the 61 patients (33%) who had persistent or increasing leakage and was successful in stopping the leakage in 17 (85%; reoperation was performed in the remaining 3 patients). TPN was the initial treatment in 15 patients (19%) and successfully stopped leakages in 11 (73%) after a median of 5 days (the remaining patients underwent reoperation). Surgical closure was the initial treatment in one patient.
• Jacobson and colleagues (2012) evaluated the impact of preoperative TPN in 15 consecutive patients with moderate-to-severe CD undergoing bowel surgery and primary anastomosis who received preoperative TPN for 18–90 days (mean, 46 days). All patients receiving TPN had clinical remission during the preoperative period, and there was no evidence of significant early postoperative complications (≤30 days), compared with 29 patients in the control group (n = 105) who did not receive TPN (p values not reported).³⁴¹

TRAUMA

• In the prospective RCT by Fan and colleagues that compared the effects of timely SPN (ETF + PN), ETF alone, and PN alone in patients undergoing surgery for severe traumatic brain injury, SPN was associated with lower rates than ETF alone for aspirated pneumonia (27.5% vs 50.0%; χ2 = 6.39, p < 0.05), hypoproteinaemia (17.5% vs 55.0%; χ2 = 18.26, p < 0.01) and diarrhoea (20.0% vs 60.0%; χ2 = 20.00, p < 0.01). Rates were lower with SPN than with PN alone for stress ulcer (22.5% vs 47.5%; χ2 = 8.24, p<0.01), intracranial infection (12.5% vs 32.5%; χ2 = 6.88, p<0.01) and pyaemia (25.0% vs 47.5%; χ2 = 6.57, p < 0.05).²⁷⁶

CANCER

• Malnutrition is associated with a high risk of early discontinuation of chemotherapy.³⁴² Pan and colleagues (2013) conducted a multicentre, cross-sectional study involving 2,248 hospitalised cancer patients in China to understand the impact of malnutrition, nutritional risk, and nutritional support on clinical outcomes.³⁴³ The rate of malnutrition at baseline and reassessment were 19.7% and 26.8%, respectively. Patients who received ETF or TPN had significantly reduced relative risk of chemotherapy-related adverse events than patients who did not receive nutrition support (RR 0.08 [95% CI 0.01–0.62] and 0.56 [0.33–0.96]).