Acta Pædiatrica ISSN 0803-5253
REVIEW ARTICLE
Enemas, suppositories and rectal stimulation are not effective in accelerating enteral feeding or meconium evacuation in low-birthweight infants: a systematic review Kim Kamphorst1,2,3, Ydelette Sietsma3,4, Annemieke J. Brouwer3,5, Paul J. T. Rood3,6, Agnes van den Hoogen (
[email protected])3,5 1.Department of Neonatology, Lucas Hospital, Apeldoorn, The Netherlands 2.Department of Gynecology, Obstetrics, Reproductive Medicine and Children, Deventer Hospital, Deventer, The Netherlands 3.Department Clinical Health Science, University Utrecht, Utrecht, The Netherlands 4.Department Viaa, Christian University of Applied Sciences, Zwolle, The Netherlands 5.Intensive Care Neonatology, Wilhelmina Children’s Hospital, University Medical Centre Utrecht, Utrecht, The Netherlands 6.Department of Intensive Care Medicine, Radboud University Medical Centre, Nijmegen, The Netherlands
Keywords Enemas, Feeding tolerance, Low-birthweight infants, Meconium excretion, Suppositories Correspondence Agnes van den Hoogen RN, PhD, Intensive Care Neonatology, Wilhelmina Children’s Hospital, University Medical Centre Utrecht, PO Box 85090, 3508 AB Utrecht, The Netherlands. Tel: +31 31 6 5523 46 48 | Email:
[email protected] Received 2 March 2016; revised 21 May 2016; accepted 8 August 2016.
ABSTRACT Early full enteral feeding in preterm infants decreases morbidity and mortality. Our systematic review covered the effectiveness of rectal stimulation, suppositories and enemas on stooling patterns and feeding tolerance in low-birthweight infants born at up to 32 weeks. It comprised seven studies published between 2007 and 2014 and covered 495 infants. Conclusion: Suppositories were ineffective in shortening the time to reach full enteral feeding, and the evidence on enemas was contradictory. Enemas and rectal stimulation did not shorten the time until complete meconium evacuation was reached. Further research into safe, effective interventions to accelerate meconium excretion is needed.
DOI:10.1111/apa.13540
INTRODUCTION Every year, approximately 15 million babies are born preterm before 37 completed weeks of gestation across the world (1). In the Netherlands, infants born before 32 weeks of gestation are admitted to a neonatal intensive care unit (NICU) to increase their chance of survival (2), and a delay in achieving full enteral feeding in these preterm infants has been closely associated with increased morbidity and mortality (3–5). A consequence of feeding intolerance is the prolonged need for total parenteral nutrition (TPN), which increases the risk of infection, liver damage and cholestasis. These complications result in prolonged hospital stays, poor neurodevelopment outcomes and mortality (4,5). Subsequently, establishing optimal enteral feeding is a priority in NICUs worldwide (3–5). In practice, premature infants start with minimal enteral feeding, but decisions about initiating and increasing feeding depend on the risk of potential complications that can be caused, especially in premature infants with low birthweight. Preterm infants with low birthweight are those who weigh 1500 g or less (6). Timing of the first
Key notes
Abbreviations NICU, Neonatal intensive care unit; RCT, Randomised controlled trial; TPN, Total parenteral nutrition; VLBW, Very low birthweight.
1280
and last meconium stool is critical for oral feeding tolerance and proper gastrointestinal function (7). Delayed passage of the first stool is very common in preterm infants (8), and the more premature the infant, the later meconium passage starts and the longer meconium passage lasts (9). Previous research has shown that low-birthweight infants with rapid meconium evacuation and a higher stool frequency during the first two weeks of life needed less time to achieve full enteral feedings (10–12). In current practice, enemas and rectal stimulation are commonly used methods to induce meconium excretion. These interventions can have an effect on the acceleration of
Early full enteral feeding in preterm infants decreases morbidity and mortality. Our systematic review covered the effectiveness of rectal stimulation, suppositories and enemas on stooling patterns and feeding tolerance in low-birthweight infants born at up to 32 weeks. We found that suppositories, rectal stimulation and enemas were not effective in shortening the time to reach full enteral feeding or complete meconium evacuation and that more research is needed.
©2016 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2016 105, pp. 1280–1287
Kamphorst et al.
meconium and thereby improve feeding tolerance, according to an observational study from Korea (13). However, a Cochrane review on the effectiveness of glycerine laxatives for the prevention or treatment of feeding intolerance in very low-birthweight (VLBW) infants did not recommend giving infants prophylactic glycerine laxatives, as these interventions did not influence the time until full enteral feeding was reached (14). A meta-analysis concluded that there was inconclusive evidence for the use of glycerine enemas or suppositories in premature infants (15). As this systematic review and meta-analysis only included randomised controlled trials (RCTs), we wondered whether there was recent evidence regarding this subject in other research designs and with no language restrictions. Rapid meconium evacuation has been identified as a key factor for feeding tolerance in low-birthweight infants (10– 12). Therefore, it is important to review the literature again, searching for all the evidence with regard to the effectiveness of interventions to accelerate meconium. The aim of this systematic review was to assess the literature for all the available evidence on the effectiveness of rectal stimulation and enemas on stooling patterns and feeding tolerance in low-birthweight infants with gestational ages of up to 32 weeks.
METHOD This systematic review was conducted according to the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (16). To explore all the relevant studies that met the inclusion criteria, the following electronic databases were systematically searched for eligible articles: PubMed, CINAHL, Embase and the Cochrane library. The electronic databases were systematically searched for ‘rectal stimulation, enema*’, ‘stooling patterns, feeding tolerance’ and ‘premature, low birth weight’ independently and in any combination. The reference lists of the relevant, included studies were screened to retrieve additional literature. The search was conducted in August 2015, and the full search strategy is presented in the Appendix S1 (Supporting Information online). Inclusion and exclusion criteria We included papers that reported on rectal stimulation, suppositories or enemas applied by nurses to prevent or treat bowel obstruction in infants of up to 1500 g and up to 32 gestational weeks and where the primary outcome measures were feeding tolerance or meconium stooling patterns. Only papers that reported on primary research were included, and there were no restrictions on date or language the paper was published in. We excluded studies using pharmacological or contrast enemas. All the records produced by the databases were screened by two reviewers (KK and YS) using the title and abstract. Full-text articles were eligible if they met the inclusion and exclusion criteria. Data on the study design setting, sample, objectives, intervention and results from
Suppositories, enemas and low birthweight
the included studies were extracted using a structured form. Methodological quality The methodological quality of the included studies was independently assessed by two researchers (KK and YS), and discrepancies in the quality assessment were resolved by discussion until consensus was reached. The Cochrane RCT checklist was used to assess the quality of the RCT study designs (17). The Cochrane evidence-based guideline development group compiled this list, based on consensus, specifically for RCT study designs, and it consists of ten items that are rated as positive, negative or unclear. A positive was given if the study fulfilled and described the criterion, negative if the study did not met the criterion and a question mark if the criterion was not described. The observational studies were rated using the Newcastle– Ottawa Scale, which was developed to assess the quality of observational studies (18). The content validity and interrater reliability of this rating scale have been established (19). The scale contains a maximum of nine points and assesses studies in three core areas: selection of study participants, comparability of groups and exposure. There are predetermined answers to the scale questions, which decide whether an outcome earns a star.
RESULTS The electronic database search identified 65 records and, after removing the duplicates, 52 records remained. The initial selection, which was based on the title and abstract, comprised 13 studies, and after reading the full texts, we excluded six papers: two systematic reviews, two studies with a different target group, one incorrect intervention and one study without an intervention. The reference lists of the seven remaining articles were screened for any new and relevant titles, but none were identified. Details of the study selection are presented in Figure S1. Five of the seven studies included in this review were RCTs (20–24), and two were observational studies (12,13). Two were published in 2007, with one each in 2011, 2012 and 2013 and the remaining two in 2014. Two of the studies were carried out in Spain, with one each from the United Kingdom, India, Chile, Korea and Austria. They covered a total of 495 infants, 300 in the intervention groups and 195 in the control groups. In two RCTs, the number of infants that were included was less than the number required by the power calculation (22,23). The methodological quality of the included studies varied, and only the case–control study fulfilled all the quality criteria (13). Four studies were rated strong: two RCTs received a score of 9/10 (22,24) and one scored 8/10 (21). Three studies were rated moderate: one study received 7/10 (23), one received 5/10 (20) and the cohort study received a score of 7/9 (12). Detailed information of the quality assessment is shown in Table S1. The frequently observed weaknesses were the lack of blinding of the
©2016 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd 2016 105, pp. 1280–1287
1281
Suppositories, enemas and low birthweight
Kamphorst et al.
Compared with the study group, less than half of the infants in the control group achieved equivalent feeding during the same period.
patients, clinicians and assessors. Extracted data from the included studies are displayed in Table 1. Two studies assessed the effect of one glycerine suppository per day (21,24). In four studies, the intervention consisted of preventive enemas until complete meconium evacuation was achieved (13,20,22,23). One study applied rectal stimulation if there was abdominal distension and if no defecation had occurred in the previous 24 hours (12). For a complete description of the different interventions, see Tables 1 and 2. In four studies, the primary outcome was the time required to achieve full enteral feeds (13,21,22,24): two described this as 100 mL/kg/day, one as 180 mL/kg/day and one had no prespecified minimum target. In one study, the primary outcome was the time when the last meconium was passed (20), and in another, it was the time when normal stooling patterns occurred (23).One study had three primary outcomes, namely postnatal age when enteral feeding volumes of at least 120 mL/kg/day were achieved, postnatal age at normal bowel habit and the age at the complete evacuation of meconium (12). Four studies had a secondary outcome of necrotising enterocolitis (13,21,22,24), three measured the episodes of culture-positive sepsis (13,21,22) and three evaluated the growth that had occurred, or postnatal age, when the birthweight was regained (13,21,24). Furthermore, the secondary outcome in two studies was the time to the first stool (13,21), and two studies used the time until a normal bowel habit was reached (12,13). Three studies measured feeding tolerance or intolerance (20,21,23,24), and three studies examined the time until full enteral feeding was reached, with two defining this as 120 mL/kg/day and one using 150 mL/kg/day (12,20,23). One study used the secondary outcome variable of ventilation days (21), one used the age when the infant was discharged from the hospital (24) and another used the introduction of oral feeding and the amount the infant was fed on the 14th day of life (20). In addition, one study examined hyperbilirubinemia, abdominal surgical exploration and intracranial haemorrhage (22).
Primary outcomes of enemas Reporting on the primary outcomes of enemas, one study found that the control group passed the last meconium five days later than the intervention group (p = 0.1) (20), while another study showed no differences between the age of completed meconium passage or normalisation of stooling patterns (23). One observational study showed significant correlations between postnatal age when 120 mL/kg/day enteral feeds were reached and postnatal age when normal stooling patterns were achieved (p < 0.0001), together with complete evacuation of meconium (p < 0.0001) and gestational age (p < 0.0001) (12). When it came to the secondary outcomes of enemas, one study showed that the suspicion of enterocolitis was more frequent in the control group than in the intervention group (p = 0.04) (22). One study showed that infants in the study group passed their first meconium faster than the control group (p < 0.001) and daily stool frequencies differed significantly in the two groups until 10 days of life, with the exceptions of days three and four (13). One study reported that gestational age was a significant predictor for last meconium passage (p = 0.02) (20). One study found that infants with significant haemodynamic patent ductus arteriosus had a delay in the attainment of full enteral feeding (p = 0.005) and normalisation of stooling patterns (p = 0.021) (23). None of the included studies showed that the interventions caused any complications.
Feeding tolerance With regard to the primary outcomes of glycerine suppositories, one study found that the time to achieve full enteral feeds was 1.6 days shorter in the intervention group, although this was not significant (21), and another found that the mean was similar in both groups (24). When it came to the secondary outcomes, one study reported as an insignificant outcome that the intervention group regained birthweight two days earlier than the control group (24). With regard to the primary outcomes of enemas, one study found that the mean time to reach full enteral feeds was similar in both groups (22). Another study found that the enemas had significant effects, reporting that the study group achieved full enteral feeding faster than the control group (p < 0.001) and that this difference became larger for infants with a birthweight of
