Archives of Disease in Childhood 1995; 72: F47-F48

F47

Exposure to invasive procedures in neonatal intensive care unit admissions

Abstract The nature and numbers of invasive procedures were studied in 54 consecutive infants admitted to a neonatal intensive care unit. Over 3000 procedures were recorded, 74% in infants below 31 weeks of gestation. One infant (23 weeks' gestation, birth weight 560 g) underwent 488 procedures. Heel prick blood sampling was the most common procedure (56%), foliowed by endotracheal suction (26%) and intravenous cannula insertion (8%). Invasive procedures which would cause pain or distress to a child are frequently performed on infants admitted to the neonatal intensive care unit. A reduction in the number of procedures, modifying them, or providing adequate analgesia could relieve some of this pain and

distress.
(Arch Dis Child 1995; 72: F47-F48)
Keywords: neonatal intensive care admissions, invasive procedures.

Sick newborn infants are subjected to a variety of invasive procedures as part of their management, some for investigation, some for treatment. In children and adults pain produced by such procedures is recognised and treated on humanitarian grounds. In the sick infant invasive procedures can cause major physiological disturbance' which can be reduced by providing analgesia,2 yet a recent survey confirmed that analgesia is used less frequently on neonatal compared with paediatric intensive care units.3 Before the provision of adequate analgesia can be addressed, a detailed examination of current practice is required. This study was designed to determine the nature and frequency of invasive procedures during neonatal intensive care.

treatment with supplemental oxygen. Thirteen (24%) were ventilated (for between one and 47 days), all but one of whom were below 31 weeks' gestation. Thirty three (80%) of the preterm babies received phototherapy, with two babies undergoing exchange transfusion. Of the 13 term babies, eight were admitted with hypoglycaemia associated with either intrauterine growth retardation or maternal diabetes. Three babies died. Two extremely premature infants of 23 and 24 weeks' gestation died aged 20 and 7 weeks, respectively, and a term baby with an abnormal karyotype died following withdrawal of intensive care at 9 days of age. Procedures were documented by the person performing them ticking a designated chart. One tick only was recorded per procedure, even if more than one attempt was made. Charts were reviewed daily and checked against nursing and medical records to ensure accurate recording. Monitoring continued as the infants progressed from intensive care to low dependency and final discharge. Most procedures surveyed involved direct tissue injury (table). Intubation and endotracheal suction were included because local trauma may occur during awake intubation, and because both cause substantial physiological disturbance.4 5 Our list of painful or noxious procedures is not exhaustive; infants undergo others not included in this study, such as gastric tube placement and removal of adhesive tape and electrodes.

Results The total number of invasive procedures recorded was 3283, of which the most common were heel prick blood sampling (56%), endotracheal suction (26%), and intravenous cannula insertion (8%) (table). The most immature infants underwent the highest numbers of procedures, with 74% performed on the 30% of infants born below 31 weeks' gestation (figure).
Numbers of procedures
Procedure
Heel prick Endotracheal suction Peripheral venous cannula Venous blood sample Intubation Intramuscular injection Venous long line Peripheral arterial line Umbilical catheter Arterial stab Lumbar puncture Chest drain Suprapubic aspiration Total

Department of Neonatal Medicine and Surgery, Nottingham City Hospital, Hucknall Road, Nottingham NGS IPB D P Barker N Rutter
Correspondence to: Dr D P Barker. Accepted 5 August 1994

Methods Fifty four consecutive admissions to the neonatal intensive care unit over a three month period were enrolled into the study. Their gestational ages ranged from 23 to 41 weeks (median 33 weeks) and their birth weights from 0f56 to 4A42 kg (median 1 9 kg). Infants transferred postnatally from other hospitals, those transferred out to other hospitals following admission, and infants admitted to the unit for less than 24 hours were excluded. Forty one (76%) infants were preterm., including 16 (30%) born at less than 31 weeks' gestation. Twenty six (48%) required

Frequency
1849 854 260 101 66 46 39 17 16 13 9 8 5 3283

F48
500

Barker, Rutter

babies depends on the difficulty of the procedure and the experience and skill of the operator. Even for heel prick sampling repeats are often required.7 The most frequently performed procedures 400 were heel prick sampling, endotracheal suction, and intravenous cannula insertion. cn Previous research has shown that the distress CD caused by heel prick sampling can be reduced D 300 by improving the method of specimen collecC) 0 Q) tion.8 The use of local anaesthetic cream 0. before skin puncture has been widely adopted 0 in paediatric practice, yet little attention has been paid to topical anaesthesia in the newE200 E born. EMIA cream is not licensed for use in z0 infants due to potential toxicity of the prilocaine component.9 However, this should not prevent a search for suitable alternatives, 100 particularly as the preterm infant becomes hypersensitive after repeated injury.10 The number of procedures performed increased dramatically in infants below 30 0 ~ ~ 00 weeks' gestation. Although this reflects both 22 26 30 34 38 42 the severity and duration of illness, in the Gestation (weeks) extreme case of the two most immature infants Number of invasive procedures v gestational age. (23 and 24 weeks' gestation), who died having undergone a combined total of over 850 A girl of 23 weeks' gestation and birth invasive procedures, this raises important weight of 560 g had the highest individual ethical issues of whether intervention on this number of invasive procedures (n=488). scale is really justified. Endotracheal suction (n= 212), heel prick Our study confirms the scope for more blood sampling (n= 218), and intravenous widespread use of analgesia on neonatal units. cannula insertion (n=29) made up over 90% Most procedures were performed without of this total. analgesia. Most ventilated babies received diamorphine by infusion, but this was usually limited to the first few days of life. Invasive Discussion procedures are necessary and unavoidable Intensive care of sick newborns has improved when caring for sick infants, but should be kept outcome, but pain and potentially adverse to a minimum. Further research is needed into physiological effects caused through procedures methods of reducing procedure related pain on and recurrent handling warrant consideration, the neonatal unit. Efforts should be concenboth on humanitarian grounds and because of trated on those procedures performed most the potential for further improving outcome. frequently, and on infants of very low birth This study forms a carefully documented weight, on whom most procedures are carried record of invasive procedures in an unselected out. group of infants admitted to our neonatal We acknowledge the assistance of staff nurse Kevin Brett during intensive care unit. The pattern and frequency this study. of invasive procedures will vary between units, 1 Speidel BD. Adverse effects of routine procedures on but is likely to be broadly similar across the preterm infants. Lancet 1978; i: 864-5. M-L. Pain relief can reduce hypoxaemia in United Kingdom. It has been suggested that 2 Pokela distressed neonates during routine treatment procedures. audit of invasive procedures should form an Pediatrics 1994; 93: 379-83. H, May A, Coates E. Use of analgesic agents for essential component of intensive care, though 3 Bauchner invasive medical procedures in pediatric and neonatal the difficulty of complete data collection has intensive care units. J Pediatr 1992; 121: 647-9. 4 Kelly MA, Finer NN. Nasotracheal intubation in the been demonstrated.6 neonate: physiologic responses and effects of atropine and The procedures in this study resulted from pancuronium. J Pediatr 1984; 105: 303-9. JM, Volpe JJ. Suctioning in the preterm infant: standard medical and nursing practices on our 5 Perlman effects on cerebral blood flow velocity, intracranial unit. We were surprised by the number of pressure, and arterial blood pressure. Pediatrics 1983; 72: 329-34. invasive procedures performed, as we do not 6 Southall DP, Cronin BC, Hartmann H, Harrison-Sewell C, consider our practice to be particularly interSamuels MP. Invasive procedures in children receiving intensive BMJ 1993; 306: 1512-3. ventional. However, our data underestimate 7 Barker DP,care. Latty BW, Rutter N. Heel blood sampling in the trauma inflicted on our patients. preterm infants - which technique? Arch Dis Child 1994; 71: F206-8. Heightened awareness among staff during the 8 Harpin VA, Rutter N. Making heel pricks less painful. Arch Dis Child 1983; 58: 226-8. study probably led to fewer procedures - for B, Nilsson A. Methemoglobinemia associated example, by improved co-ordination of blood 9 Jakobson with a prilocaine-lidocaine cream and trimetoprimsampling - and the number of attempts per sulphamethoxazole. A case report. Acta Anaesthesiol Scand 1985; 29: 453-5. procedure was not recorded. The actual 10 Fitzgerald M, Millard C, McIntosh N. Hyperalgesia in number of skin punctures experienced by premature infants. Lancet 1988; i: 292.
.
_