Randomised trial of umbilical arterial position : Doppler ultrasound findings

Umbilical arterial catheters (UAC) were randomly assigned in 69 infants to a high (n=36) or to a low (n=33) position. Serial Doppler ultrasound measurements of blood flow velocity in their superior mesenteric arteries, coeliac axis, renal arteries, and anterior cerebral arteries were then obtained. There were no differences in blood flow velocity between high and low UAC groups on days 1, 3, and 7. At 2 weeks, those infants with a high UAC still in place had significantly higher velocities in the mesenteric artery than those infants who had no catheter in place. Infants with high UACs remaining in place for more than 7 days were found to have an increase in abdominal distension and tenderness, whereas this was not the case for those with low UACs. Catheter position has no effect on visceral blood flow if the UAC stays in place for one week or less, whereas prolonged use of a high UAC may alter intestinal blood flow and increase the incidence of abdominal symptoms. Children Nationwide Neonatal Centre, King's College Hospital, London S T Kempley H R Gamsu Correspondence to: Dr S T Kempley, Frederick Still Ward, King's College Hospital, London SE5 9RS. Accepted 7 January 1992 There has been a longstanding controversy concerning the role of umbilical arterial catheters (UAC) in the aetiology of necrotising enterocolitis. Case-control studies have produced conflicting results,1-'2 with the suggestion that UACs may have an independent aetiological role in some birthweight groups while their use is merely associated with other definitive risk factors in others.12 If it is assumed that UACs do have a truly causative role, by compromising blood flow in the mesenteric arteries, then catheters placed so that the tip lies above the origin of these arteries (high UACs) would be expected to cause more necrotising enterocolitis than catheters in a low position. However, randomised studies have failed to find any excess of necrotising enterocolitis in those infants with high catheters,13 14 a finding confirmed in a large randomised study that we have conducted (ST Kempley, S Bennett, BG Loftus, D Cooper, HR Gamsu, unpublished observations). This is surprising in view of the fact that catheter associated thrombosis has been found in a high proportion of infants with a UAC,1314 and clinical involvement of the mesenteric arteries was documented principally with high UACs. However, the majority of aortic thrombi are clinically silent and it is possible catheter that catheters, or catheter associated thrombosis, could affect intestinal blood flow without necessarily causing necrotising enterocolitis. We have therefore used Doppler ultrasound to document the effects of UAC position on blood flow velocity in the superior mesenteric artery (SMA), coeliac axis, and left renal artery. These major branches of the abdominal aorta originate proximally to the tip of UACs placed in a low position, but distal to the tip of UACs placed high; blood flow in them should therefore only be influenced in the latter. We have also included measurements on the anterior cerebral artery (ACA) as an index of each infant's general cardiovascular status. Methods Over a two year period 147 infants admitted to our unit and requiring umbilical arterial catheterisation had their UAC randomly allocated to a high position (T6-T10) or a low position (below L3). Catheters were inserted at a median age of 2 hours to a distance determined from the shoulder-umbilicus length and the position was adjusted after a plain radiograph. The catheters were either of a plain end hole type (Argyle 3-5 French gauge (FG) or Argyle 5 0 FG, Sherwood Medical) or with a side hole and an oxygen sensing electrode at the tip (Neocath 4 0 FG or Neocath 5 0 FG, Biomedical Sensors Ltd). The initial infusate was 5°/O dextrose with 1 IU/ml heparin, run at 1 ml/hour in all infants. Blood pressure monitoring was performed via the UAC in all infants. A birthweight stratified randomisation was used, with permuted blocks within strata, to help produce well matched groups. One hundred and tiirty one infants had their catheter successfully inserted into the correct, randomised position. (Over this time period 36 infants had a UAC in place but were not randomised; most of these already had a UAC at the time of transfer from another hospital, or had been considered clinically unsuitable for randomisation before entering the trial.) In order to be included in the study infants were required to have a UAC placed in the position to which they had been randomised and to have had Doppler measurements within the first 24 hours of life.. By excluding data from infants who had their first measurement after 24 hours we ensured that infants were not selected for study on the basis of their later clinical course, and so avoided biased groups. The final study group consisted of 69 infants, 855 group.bmj.com on April 13, 2017 Published by http://adc.bmj.com/ Downloaded from