Systematic review and meta-analysis of use of serum C-reactive protein levels to predict anastomotic leak after colorectal surgery

BACKGROUND

Several recent studies have investigated the role of C-reactive protein (CRP) as an early marker of anastomotic leakage following colorectal surgery. The aim of this systematic review and meta-analysis was to evaluate the predictive value of CRP in this setting.

METHODS

A systematic literature search was performed using MEDLINE, Embase and PubMed to identify studies evaluating the diagnostic accuracy of postoperative CRP for anastomotic leakage following colorectal surgery. A meta-analysis was carried out using a random-effects model and pooled predictive parameters were determined along with a CRP cut-off value at each postoperative day (POD).

RESULTS

Seven studies, with a total of 2483 patients, were included. The pooled prevalence of leakage was 9·6 per cent and the median day on which leakage was diagnosed ranged from POD 6 to 9. The serum CRP level on POD 3, 4 and 5 had comparable diagnostic accuracy for the development of an anastomotic leak with a pooled area under the curve of 0·81 (95 per cent confidence interval 0·75 to 0·86), 0·80 (0·74 to 0·86) and 0·80 (0·73 to 0·87) respectively. The derived CRP cut-off values were 172 mg/l on POD 3, 124 mg/l on POD 4 and 144 mg/l on POD 5; these corresponded to a negative predictive value of 97 per cent and a negative likelihood ratio of 0·26-0·33. All three time points had a low positive predictive value for leakage, ranging between 21 and 23 per cent.

CONCLUSION

CRP is a useful negative predictive test for the development of anastomotic leakage following colorectal surgery.

Ethnic disparity in colonic cancer outcomes in New Zealand--biology or an access issue?

BACKGROUND

There is an ethnic variation in outcomes for colonic cancer in New Zealand. Whether this disparity is caused by cancer biology or inequitable provision of treatment services after diagnosis has not been elucidated.

METHOD

National cancer registry data from 1996 to 2003 were obtained. Incidence and mortality rates for the four major ethnic groups were age-adjusted to the new WHO world population. The impact of age, sex, AJCC stage and site of cancer at diagnosis was compared between ethnic groups using a Cox regression analysis.

RESULTS

A total of 11 987 colonic cancer registrations were identified. The overall raw 5-year mortality was 53.7%. The age-adjusted incidence in Europeans was more than double that of the Maori, Asian and Pacific populations at 33.0 per 100,000 population/year. Europeans presented at a greater age, with more right sided cancers, and at an earlier stage of disease. The opposite was true for the Maori population. Pacific Islanders and Asians presented at a younger age, but with a similar site, stage and sex distribution to the rest of the population. There were no significant differences in 5 year mortality after diagnosis when age, sex, stage, and site at presentation were controlled for by cox regression analysis.

CONCLUSION

These results suggest that age, sex, stage and site at presentation may be more important than inequality in treatment provision after diagnosis in explaining differences in outcomes between the ethnicities. Efforts need to be focused on identifying reasons for the increased risk of colonic neoplasia in Europeans and the later stage disease presentation in the Maori population.

Randomized clinical trial of the effect of glucocorticoids on peritoneal inflammation and postoperative recovery after colectomy

Background

Recent data have suggested a relationship between postoperative fatigue and the peritoneal cytokine response after surgery. The aim of this study was to test the hypothesis that preoperative administration of glucocorticoids before surgery would decrease fatigue and enhance recovery, by reducing the peritoneal production of cytokines.

Methods

In a double-blind randomized controlled study, patients undergoing elective, open colonic resection were administered 8 mg dexamethasone or normal saline. Patients were treated within an enhanced recovery after surgery programme. Primary outcomes were cytokine levels in peritoneal drain fluid and fatigue as measured by the Identity–Consequence Fatigue Scale (ICFS).

Results

Baseline parameters were similar for 29 patients in the dexamethasone group and 31 in the placebo group. Patients who received dexamethasone had lower ICFS scores on days 3 and 7. Dexamethasone was associated with significantly lower peritoneal fluid interleukin (IL) 6 and IL-13 concentrations on day 1, and these correlated with changes in the ICFS score. There was no significant increase in adverse events in the dexamethasone group.

Conclusion

Preoperative administration of dexamethasone resulted in a significant reduction in early postoperative fatigue, associated with an attenuated early peritoneal cytokine response. Peritoneal production of cytokines may therefore be important in postoperative recovery. Registration number: ACTRN12607000066482 (http://www.anzctr.org.au/).

Randomized trial of oral sodium phosphate compared with oral sodium picosulphate (Picolax) for elective colorectal surgery and colonoscopy

BACKGROUND

Sodium picosulphate (Picolax) is considered by most British surgeons as standard preparation for colonoscopy and elective surgery. Oral sodium phosphate may be better tolerated and more efficient as bowel preparation.

METHODS

A randomized trial was performed to compare oral sodium phosphate (n = 76) with Picolax (n = 77) as bowel preparation for elective colorectal surgery. A parallel study randomized colonoscopy patients to sodium phosphate (n = 51) or Picolax (n = 52). Patient acceptability was measured for seven symptoms with a linear analogue score. Quality of preparation was graded by the surgeon and faecal residue was measured in resection specimens. During colonoscopy, bowel preparation has graded 0-24 using an endoscopic score.

RESULTS

Abdominal pain, nausea, vomiting, embarrassment, fear and fatigue did not differ significantly between the groups. Surgeons grade of quality was judged poor or awful in 5 of 76 in the sodium phosphate group (9%) compared with 13 of 73 in the Picolax group (18%, p = 0.084). Mean faecal residue in the resection specimen was 0.1 g/cm after sodium phosphate compared with 0.45 g/cm after Picolax (p < 0.01). The endoscopic score was significantly lower using sodium phosphate (2.0 +/- 2.2) than picolax (3.1 +/- 2.9; p < 0.05).

CONCLUSIONS

These results suggest that oral sodium phosphate is well tolerated and superior to Picolax in elective colorectal surgery and colonoscopy.

Manipulations of the metabolic response for management of patients with severe surgical illness: review

The metabolic response to severe surgical illness is complex and varied. Much recent laboratory and clinical research has focused on increasing our understanding of the metabolic response and the development of new therapies designed to modify this response. Antiinflammatory agents can target harmful aspects of the metabolic response; the immune system can be stimulated; and anabolic factors can be used in an attempt to enhance recovery. The nutritional support of the surgical patient remains crucial, but the effects of new additives are being studied in a variety of surgical conditions. As yet, few of these "novel" agents have found an established role in the management of surgical patients. This review focuses on many "novel" agents or those that do not yet have a clearly defined role in surgical illness. Clinical trials in the areas of severe sepsis, major surgical trauma, and major elective surgery have been emphasized.

Outcome of strictureplasty for duodenal Crohn's disease

BACKGROUND

The outcome of strictureplasty for duodenal Crohn's disease has not been critically documented. The aim of this study was to assess the outcome of strictureplasty for duodenal Crohn's disease.

METHODS

A retrospective review was undertaken of 13 patients who underwent strictureplasty (including four pyloroplasties) for obstructive duodenal Crohn's disease between 1974 and 1997.

RESULTS

Ten patients underwent strictureplasty as the primary procedure, and in three strictureplasty was used as a revision procedure after previous bypass surgery. Two patients developed anastomotic breakdown and were treated either by Roux-en-Y duodenojejunostomy or partial gastrectomy. Symptoms of obstruction persisted in four patients after strictureplasty; three eventually resolved after prolonged nasogastric aspiration, but the other required gastrojejunostomy. In the long term, six patients developed restricture at the previous strictureplasty site. Five required repeat strictureplasty and the other patient underwent duodenojejunostomy. One patient who had repeat strictureplasty required a further strictureplasty because of restricture at the previous strictureplasty site. Overall nine of 13 patients required further surgery because of early postoperative complications or restricture at the strictureplasty site.

CONCLUSION

Strictureplasty for duodenal Crohn's disease is associated with a high incidence of postoperative complications and restricture.

Effect of a chimeric antibody to tumor necrosis factor-alpha on cytokine and physiologic responses in patients with severe sepsis--a randomized, clinical trial

OBJECTIVES

Tumor necrosis factor (TNF)-alpha appears central to the pathogenesis of severe sepsis, but aspects of the cytokine cascade and the link to physiologic responses are poorly defined. We hypothesized that a monoclonal antibody to TNF-alpha given early in the course of severe sepsis would modify the pattern of systemic cytokine release and, as a consequence, resuscitation fluid requirements, net proteolysis, and hypermetabolism would be reduced.

DESIGN

Randomized, double-blind, placebo-controlled trial.

SETTING

Critical Care Unit and University Department of Surgery in a single tertiary care center.

PATIENTS

Fifty-six patients (from 92 eligible patients) with severe sepsis. Twenty-eight patients were randomized to treatment, and were comparable with the placebo group for age, gender, race, Acute Physiology and Chronic Health Evaluation II score, and site and type of infection.

INTERVENTIONS

A 300-mg single dose of cA2 (a chimeric neutralizing antibody to TNF-alpha) was given intravenously within 12 hrs of the onset of severe sepsis. Standard surgical and intensive care therapy was otherwise delivered.

MEASUREMENTS AND MAIN RESULTS

Plasma concentrations of TNF-alpha, interleukin (IL)-1beta IL-6, IL-8, IL-10, soluble 75-kilodalton TNF-alpha receptor (sTNFR-75), and IL-1beta receptor antagonist (IL-1ra) were measured by sandwich enzyme-linked immunosorbent assay before cA2 infusion, 8 hrs later, and then daily for a minimum of 4 days. Sequential changes in total body protein, body water spaces, and resting energy expenditure over 21 days were measured, as soon as patients achieved hemodynamic stability, by in vivo neutron activation analysis, tritium and bromide dilution, and indirect calorimetry, respectively. Twenty-one patients died, ten having received cA2. Suppression of measurable TNF-alpha was observed at 8 hrs with subsequent rebound by 24 hrs after cA2 treatment. The concentrations of other cytokines were high, were not reduced by intervention, and decreased logarithmically over 5 days. Both groups reached hemodynamic stability at similar times (57.5 +/- 11.8 hrs in controls vs. 58.6 +/- 9.2 hrs in the cA2 group) and following similar volumes of infused fluids (29.1 +/- 3.4 L vs. 28.9 +/- 4.4 L). No differences in net proteolysis, resolution of body water expansion, or alteration in resting energy expenditure were demonstrated.

CONCLUSION

A single dose of cA2 did not alter the overall pattern of cytokine activation or the profound derangements in physiologic function that accompany severe sepsis.

Gastroduodenal fistulas in Crohn's disease: clinical features and management

PURPOSE

The aim of this study was to assess the clinical features and management of fistulas involving the stomach and duodenum (gastroduodenal fistulas) in patients with Crohn's disease.

METHODS

The medical records of 14 patients with a gastroduodenal fistula complicating Crohn's disease treated in this unit between 1958 and 1997 were reviewed.

RESULTS

In six patients a gastroduodenal fistula was diagnosed before surgery, whereas eight gastroduodenal fistulas were discovered during surgery for distal Crohn's disease. In six patients, the fistula originated from Crohn's disease in the transverse colon, and in six patients, it originated from a recurrent disease at an ileocolonic anastomosis; these patients had no gross evidence of gastroduodenal Crohn's disease. In one patient, the ileocolonic-duodenal fistula closed on medical treatment. The other 11 patients underwent resection of the diseased bowel and closure of the gastric or duodenal fistulas. The two remaining fistulas were from the duodenum to the abdominal wall; both had primary Crohn's duodenitis. One duodenocutaneous fistula was treated by debridement of the duodenal fistula and simple closure of the defect; the other was treated by limited duodenal excision around the fistula and by duodenojejunostomy. In all patients, gastroduodenal fistulas were cured, and there have been no fistula recurrences.

CONCLUSIONS

Simple closure of the gastroduodenal component of the fistula is generally advised for gastroduodenal fistulas. However, when the duodenal defect after excision around the fistula is large, duodenojejunostomy is recommended, provided there is no evidence of jejunal Crohn's disease.

Sequential changes in the metabolic response in severely septic patients during the first 23 days after the onset of peritonitis

OBJECTIVE

To quantify the sequential changes in metabolic response occurring in patients with severe sepsis after the onset of peritonitis.

SUMMARY BACKGROUND DATA

Understanding the changes in energy expenditure and body composition is essential for the optimal management of severely septic patients; however, they have not been quantified in the context of modern surgical care.

METHODS

Twelve patients with severe sepsis secondary to peritonitis (median APACHE II score = 21.5) had measurements of energy expenditure and body composition as soon as they were hemodynamically stable and 5, 10, and 21 days later. Sequential measurements of acute-phase proteins and cytokine responses were also made.

RESULTS

Resting energy expenditure rose to 49% above predicted and remained elevated throughout the study period. Total energy expenditure was 1.25 x resting energy expenditure. Body fat was oxidized when energy intake was insufficient to achieve energy balance. There was a positive fluid balance of 12.5 1 over the first 2 days after onset of sepsis; thereafter, body water changes closely paralleled body weight changes and were largely accounted for by changes in extracellular water. During the 21 -day study period, there was a loss of 1.21 kg (13%) of total body protein. During the first 10 days, 67% of the protein lost came from skeletal muscle, but after this time it was predominantly from viscera. Intracellular potassium levels were low but did not deteriorate further after hemodynamic stability had been reached. There was a reprioritization of hepatic protein synthesis that was obligatory and independent of changes in total body protein. The cytokine responses demonstrated the complexity, redundancy, and overlap of mediators.

CONCLUSIONS

The period of hypermetabolism in severely septic patients is similar to that previously described, but the fluid changes are larger and the protein loss is greater. Protein loss early on is predominantly from muscle, thereafter from viscera. Fat loss can be prevented and cell function preserved once hemodynamic stability is achieved.

Results of restorative proctocolectomy in the elderly

Twenty eight of 227 patients undergoing restorative proctocolectomy for inflammatory bowel disease, familial adenomatous polyposis or functional disease were over the age of 50 years: ages 50 to 60 (n = 13), 60 to 70 (n = 10), and over 70 (n = 5). Major complications occurred in 5 patients over the age of 50 (18%) compared with 43 patients under the age of 50 (23%). Three patients above the age of 50 had their pouch excised (11%) compared with 23 under the age of 50 (12%). Functional outcome was assessed with a 12 point symptom score. This was similar in all age bands: under 50 years (mean = 2.2; sd +/- 2.2; n = 109), 50 to 60 years (mean = 2.5; sd +/- 2.5; n = 12), 60 to 70 years (mean = 2.8; sd +/- 2.3; n = 7) and over 70 years (mean = 4.0; sd +/- 3.7; n = 5): P > 0.05). When analysed for ulcerative colitis alone, no significant differences were seen between the two age groups. Restorative proctocolectomy in the elderly gives results which are comparable to the younger population.

Quality of life after pouch excision

BACKGROUND

Pouch excision is a devastating experience for patients having restorative proctocolectomy for ulcerative colitis.

METHODS

The quality of life among patients having pouch excision for ulcerative colitis was compared with that in those having proctocolectomy and ileostomy for ulcerative colitis using a validated standardized self-administered questionnaire.

RESULTS

After pouch excision patients (n = 9) had more troublesome bowel symptoms (mainly from liquid stoma output) than those in the proctocolectomy group (n = 14) (mean(s.d.) score 5.64(0.92) versus 6.13(0.37), P = 0.03). However, the mean scores for the other parameters (systemic symptoms, functional, social and emotional impairment) did not differ significantly.

CONCLUSION

Patients having pouch excision for ulcerative colitis have more liquid ileostomy loss but a comparable quality of life to those treated by standard proctocolectomy and ileostomy.

Assessment of involuntary muscle function in patients after critical injury or severe sepsis

BACKGROUND

Study of involuntary skeletal muscle function (MFA) has been well accepted in the area of nutrition assessment and potentially offers a means for following progress of the critically ill patient. We report on the application of this technique to intensive care patients.

METHODS

MFA was performed by study of the contraction/relaxation characteristics of the adductor pollicis muscle of the thumb after ulnar nerve stimulation. Serial measurements were made in 16 critically injured patients and 28 patients with severe sepsis and were compared with those obtained from 26 control subjects. Extent of loss of total body protein (TBP) was quantified with in vivo neutron activation.

RESULTS

Significant difficulties exist in applying this technique to intensive care patients. In the critically injured, only five acceptable traces could be obtained from a possible 58 measurements. For patients with severe sepsis it was possible to obtain an acceptable trace on 12 of 56 occasions. Neuromuscular blockade and lack of patient cooperation were significant impediments to MFA study. Although frequently perceived as unpleasant by these patients, there was no long-term morbidity associated with MFA. No significant differences were seen in maximal relaxation rate at 30 Hz (MMR30) or force frequency ratios (F10/50 and F30/ 50) between trauma patients and controls. In the sepsis patient group, a significantly higher F10/50 was measured (52% +/- 3% severe sepsis vs 40% +/- 1% control subjects, p < .01). Six patients had MFA measured approximately 21 days after the illness, by which stage they had lost 11% of their initial TBP. Compared with control subjects, no significant differences were observed in MRR30 or F30/50, whereas a higher value for F10/50 was measured (48% +/- 1% critical illness vs 40% +/- 1% control subjects, p < .01).

CONCLUSIONS

The MFA technique is difficult to apply to intensive care patients. No significant disturbance to MFA is seen after critical injury. Severe sepsis results in an elevation of F10/ 50 only. When able to be obtained, MFA results do not reflect the extent of proteolysis but are indicative of the state of cellular energetics.

Progressive cellular dehydration and proteolysis in critically ill patients

BACKGROUND

According to a recent hypothesis, the profound loss of body protein that occurs in critically ill patients is triggered and maintained by cell shrinkage secondary to cellular dehydration. We tested this hypothesis by studying sequential changes in intracellular water, total body protein, total body potassium, and intracellular potassium in patients receiving intensive care for blunt trauma or sepsis.

METHODS

Nine patients with multiple blunt trauma and 11 with severe sepsis were studied in an intensive care unit for 21 days. Intracellular water was measured in two ways--by subtraction of extracellular water (bromide dilution) from total body water (tritium dilution), and by bioimpedance spectroscopy. Total body protein was measured by whole-body neutron activation analysis and total body potassium by whole-body counting.

FINDINGS

Over the study period intracellular water decreased by 15-20%, total body protein by 15%, and total body potassium by about 20%. Intracellular potassium concentration did not change, and was similar to that in healthy adult volunteers. In the trauma patients, sequential measurements of the ratio of potassium to protein in lost tissue indicated that cells were losing water in quantities greater than would be expected from protein losses.

INTERPRETATION

The loss of protein and potassium from body stores in major trauma or sepsis is accompanied by progressive cellular dehydration. This insight opens up new therapeutic options for limiting the loss of body protein in critically ill patients.