The intestinal lesions and circulating lysosomal enzymes in extracorporeal circulation

Routine enteral nutrition in neonates on extracorporeal membrane oxygenation

OBJECTIVES

To evaluate over a 5-yr period the feasibility and tolerance of a protocol of routine enteral nutrition in neonates requiring extracorporeal membrane oxygenation (ECMO).

DESIGN

Retrospective medical chart review.

SETTING

Level III children's hospital, pediatric surgical intensive care unit.

PATIENTS

Neonates treated with venoarterial ECMO (VA-ECMO) between January 1997 and January 2002. Patients with congenital diaphragmatic hernia were excluded.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Charts of all neonates treated with VA-ECMO were reviewed. Feasibility was evaluated by recording the time period needed for enteral nutrition to reach 40% of total fluid intake; tolerance was evaluated by reviewing data on enteral nutrition related morbidity. Sixty-seven of the 77 eligible patients received enteral feeding during ECMO. Thirty-six of these patients (54%) received 40% of total fluid intake as enteral nutrition within a median of 3 (range, 2-4) days. Over the years there was a trend toward an increasing usage of enteral nutrition from 71% to 94% (p = .07). Enteral nutrition was temporarily discontinued in 16 patients, with 14 showing gastric retentions, one showing discomfort, and one showing aspiration. Symptoms of bilious vomiting, blood-stained stool, or abdominal distention were not present.

CONCLUSION

Neonates on ECMO in this series tolerated enteral feeding well and did not show serious adverse effects. Overall, it is our experience that routine use of enteral feeding in critically ill neonates on VA-ECMO is feasible.

Gut hormone profiles in critically ill neonates on extracorporeal membrane oxygenation

OBJECTIVES

The objective of this study was to gain insight into the hormonal responses to enteral nutrition in critically ill newborns requiring venoarterial extracorporeal membrane oxygenation (ECMO) by analyzing plasma gut hormone levels of gastrin, cholecystokinin and peptide-YY in relation to enteral nutrition.

METHODS

In 24 consecutive neonates treated with venoarterial ECMO intestinal hormone secretions were determined by radioimmunoassay at 2-day intervals. Twelve received parenteral nutrition only. In 12 enteral nutrition was introduced later. The findings in these patients were compared with those of 16 measurements in eight non-ECMO treated age-matched controls. Mixed model analysis of variance was used for statistical analysis.

RESULTS

Concentrations of gastrin, cholecystokinin and peptide-YY were significantly higher in ECMO patients receiving enteral nutrition compared with ECMO patients who received parenteral nutrition (62, 3.8 and 59.4 pmol/L versus 46, 3.1 and 34.7 pmol/L, respectively). Overall, plasma hormone levels did not differ from those in age-matched controls.

CONCLUSIONS

Intestinal hormone levels showed normal responses after introduction of enteral feeding, comparable with those in age-matched controls without ECMO. These results do not provide an argument for withholding enteral nutrition even in the most severely ill neonates on venoarterial ECMO.

Predictors for infection following open-heart surgery

A large open-heart surgery unit dealing with approximately 1000 elective patients per annum and housed in a dedicated suite, including a specialized intensive therapy unit, has provided a unique model for the study of predictors of infections and related phenomena. Over more than 10 years a series of studies has shown: the powerful effect of antibiotic usage on the colonization/infection rates with Gram-negative bacilli; the predictive relationships between preoperative chest status and the development of postoperative chest problems; an important correlation between preoperative peak urine flow rates and susceptibility to postoperative urinary tract infections; and the predictive significance of preoperative antibodies to enteric Gram-negative bacilli in relation to postoperative pyrexia and Gram-negative infection. The lessons learned from this easily studied group of patients may have important applications in other surgical settings.

Suppressive effect of ulinastatin on plasma fibronectin depression after cardiac surgery

Plasma fibronectin, an opsonic glycoprotein, is known to modulate the reticuloendothelial phagocytic clearance of nonbacterial and, possibly, bacterial particulates. The decreased plasma fibronectin levels seen after cardiac surgery have been considered to derive mainly from opsonic consumption. In the present study, we demonstrated that the administration of ulinastatin, a human urinary trypsin inhibitor, to patients after cardiac surgery not only inhibited the postoperative depression of plasma fibronectin levels, but also maintained the plasma fibronectin level within the normal range. This effect apparently resulted from the inhibitory activity of ulinastatin on the proteolytic enzymes released after operation. This result suggests that the decreased plasma fibronectin level noted after cardiac surgery may derive mostly from excessive proteolytic enzymes. Our observation also indicates that the prophylactic administration of ulinastatin to patients undergoing major operations will result in a favorably functional reticuloendothelial phagocytic system.

Abdominal complications following cardiopulmonary bypass in open-heart surgery

Open-heart surgery was performed on 1686 adult patients between 1980 and 1984, with a mortality rate due to abdominal complications of 0.6%. Every operation involved the use of cardiopulmonary bypass (CPB). Abdominal complications occurred in 1.6%, with an overall mortality rate of 36%. The most frequent complication was gastrointestinal (Gl) bleeding (61%). The mortality of the patients who bled from the Gl tract was 53%. Other complications encountered were gastroduodenal ulcer, colitis, ileus, subphrenic abscess and intraperitoneal bleeding. Prolonged CPB and low output syndrome preceded multiple organ failure, which occurred in 39% of those who had abdominal complications and in 59% of those who bled. Gl bleeding after CPB did not correlate with a previous history of gastric ulcer. Reoperation because of cardiac tamponade or excessive chest tube drainage was a factor predisposing to Gl bleeding. The mortality and abdominal complication rates were significantly higher in valve surgery than in coronary revascularization.