Routine Parenteral Nutrition Research Articles

A Minimally Interventional Approach to Oesophageal Atresia Repair With Early Enteral Feeding is Safe, Optimises Neonatal Outcomes, and Reduces Resource use

PurposeRecent series of newborn Oesophageal Atresia (OA) repair continue to report widespread use of chest drains, gastrostomy, routine contrast studies and parenteral nutrition (PN) despite evidence suggesting these are superfluous. We report outcomes using a minimally interventional approach to post-operative recovery. MethodsEthically approved (15/WA/0153), single-centre, retrospective case-note review of consecutive infants with OA 2000–2022. Infants with OA and distal trache-oesophageal fistula undergoing primary oesophageal anastomosis at initial surgery were included (including those with comorbidities such as duodenal atresia, anorectal malformation and cardiac lesions). Our practice includes routine use of a trans-anastomotic tube (TAT), no routine chest drain nor gastrostomy, early enteral and oral feeding, no routine PN and no routine contrast study. Data are median (IQR). ResultsOf total 186 cases of OA treated during the time period, 157 met the inclusion criteria of which 2 were excluded as casenotes unavailable. TAT was used in 150 infants. A chest drain was required in 13 (8%) and two infants had a neonatal gastrostomy. Enteral feeds were started on postoperative day 2 (2–3), full enteral feeds established by day 4 (4–6) and oral feeds started on day 5 (4–8). PN was required in 15%. Median postoperative length of stay was 10 days (8–17). Progress was quicker in term infants than preterm. One infant died of cardiac disease prior to neonatal discharge. Two planned post-operative contrast studies were performed (surgeon preference) and a further 7 due to clinical suspicion of anastomotic leak. Contrast study was therefore avoided in 94%. There were 2 anastomotic leaks; both presented clinically at day 4 and day 8 after oral feeds had been started. ConclusionOur minimally interventional approach is safe. It facilitates prompt recovery with lower resource use, reduced demand on nursing staff, reduced radiation burden, and early discharge home compared to published series without adversely affecting outcomes. Level of EvidenceLevel 4.

Is intravenous fish oil associated with the neurodevelopment of extremely low birth weight preterm infants on parenteral nutrition?

Preterm infants are at increased risk of long-term neurodevelopmental disabilities (NDD). Long chain n-3 fatty acids play a key role during the development of the central nervous system and some studies in preterm infants showed benefits of docosahexaenoic acid and arachidonic acid supplementation for visual and cognitive development. In recent years fish oil has been added to the fat blend of intravenous (IV) lipid emulsions (LE) but to date scanty data are available on neurodevelopmental outcome of preterm infants that received fish oil containing LE. We studied the effect of fish oil containing IV LE vs standard IV LE on neurodevelopment in a large cohort of preterm infants who received routine parenteral nutrition (PN) from birth. We retrospectively reviewed the neurodevelopmental outcome of 477 preterm infants (birth weight (BW): 400-1249g and gestational age (GA) at birth: 24+0 - 35+6 weeks (W)) admitted to our NICU between Oct-2008 and June-2017, who received routine PN with different LE, with and without fish oil (IV-FO vs CNTR). We compared neurodevelopment at 2 years corrected age by the Bayley III development scale and the incidence of NDD. Demographics, birth data and the incidence of the main clinical short-term outcomes of prematurity were similar in the two groups (IV-FO: n=178, GA 197±14 days, BW 931±182g; CNTR: n=192, GA 198±15 days, BW 944±194g). No differences were found in maternal demographics nor in parental education between the two groups. Cognitive score was not significantly different between IV-FO and CNTR (92±15 vs 93±13, p=0.5). No differences were found in motor and language scores, and in the incidence of NDD in the two groups. Contrary to our hypothesis, the use of fish oil containing LE in a large cohort of preterm infants on routine PN did not result in better neurodevelopment. Large randomized controlled trials powered for neurodevelopment are needed to clarify the impact of the widely used fish oil containing LE on neurodevelopment of preterm infants.

Hypertriglyceridemia and lipid tolerance in preterm infants with a birth weight of less than 1250 g on routine parenteral nutrition

To study the association of hypertriglyceridemia and of lipid tolerance with clinical and nutritional data in preterm infants receiving routine parenteral nutrition. We retrospectively studied 672 preterm infants (gestational age <32 weeks) with birth weight <1250g, consecutively admitted to our NICU, born between 2004 and 2018. Selected prenatal data and interventions, parenteral intakes and diseases were considered. Hypertriglyceridemia was defined as plasma triglycerides >250mg⋅dL-1. Lipid tolerance was defined as the ratio of plasma triglycerides to the intravenous lipid intake at the time of sampling. Variables associated to hypertriglyceridemia and to lipid tolerance were identified by multiple logistic and linear regression analyses. Hypertriglyceridemia occurred in 200 preterm infants (30%), ranging from 67% at 23 weeks to 16% at 31 weeks' gestation. In 138 infants (69%) hypertriglyceridemia occurred at a lipid intake of 2.5g⋅kg-1 or less. Lipid tolerance was reduced especially in infants of less than 28 weeks' gestation (14.3±9.3 vs 18.8±10.2, respectively, p<0.001). Lipid tolerance was negatively associated with respiratory distress syndrome (OR=-1.14, p=0.011), patent ductus arteriosus (OR=-1.73, p<0.001), small for gestational age (OR=-2.96, p<0.001), intraventricular haemorrhage (OR=-3.96, p<0.001), late onset sepsis (OR=-8.56, p=0.039). Preterm infants on routine parenteral nutrition were able to tolerate markedly lower intravenous lipid intakes than the recommended target values of current guidelines. Lipid tolerance was associated with some of the major complication of prematurity, possibly at risk of developing hypertriglyceridemia.

Oxygen saturation to fraction of inspired oxygen ratio in preterm infants on routine parenteral nutrition with conventional or fish oil containing lipid emulsions.

The benefits of intravenous (IV) fish oil (FO), as a source of n-3 long-chain polyunsaturated fatty acids, on lung growth in preterm infants, remain controversial. To evaluate if IV FO improves lung growth in small preterm infants on routine parenteral nutrition (PN). We retrospectively reviewed prospectively collected data of preterm infants with a birth weight <1250 g who received routine PN from birth. We compared patients who received FO containing IV lipid emulsions with infants who received conventional emulsions (CNTR). The oxygen saturation (SpO2 ) to a fraction of inspired oxygen (FiO2 ) ratio (SFR) at 36 weeks (W) of gestation was chosen as the primary outcome variable to assess lung growth. Four hundred and seventy-seven infants were studied: 240 received IV FO and 237 CNTR. While exposure to antenatal glucocorticoids was higher in IV FO group than in CNTR (95 vs 90%, P = .04), there were no differences in birth data, enteral and parenteral nutrition intakes, ventilator supports and drug therapies. The incidence of the most common complications of prematurity at 36 W was not different (bronchopulmonary dysplasia was 27 vs 21% in IV FO vs CNTR infants, P = .1). Weight gain from birth to 36 W was marginally, but significantly, higher (+0.5 g/kg/d, P = .03) in IV FO group vs CNTR. SFR increased from 32 W to 36 W in all study patients (P < .001). IV FO infants had significantly lower SpO2 from 33 W to 35 W (P < .001) and lower (worse) SFR at 36 W (432 ± 57 vs 444 ± 51, P = .026) compared to CNTR. Contrary to our hypothesis, the use of FO containing IV lipid emulsions for the routine PN of the preterm infant did not improve lung growth compared to the infants who received conventional IV lipid emulsions.

Hypertriglyceridemia and Intravenous Lipid Titration During Routine Parenteral Nutrition in Small Preterm Infants.

In case of hypertriglyceridemia (HiTG) during parenteral nutrition (PN), the 2018 European Society of Paediatric Gastroenterology, Hepatology and Nutrition guidelines recommend an intravenous (IV) lipid titration, but its consequences in small preterm infants are largely unknown. We compared macronutrient and energy intakes, growth, diseases associated with prematurity, and neurodevelopment in small preterm infants on PN who developed (cases) or did not develop HiTG (controls, CNTR). We retrospectively reviewed data of preterm infants with a birth weight (BW) <1250 g consecutively admitted to our neonatal intensive care unit (2004-2016) who received routine PN. HiTG infants were defined by at least 1 triglyceride (TG) measurement >250 mg/dL during the first 10 days of life. Patients with and without HiTG were match-paired for BW and gestational age. A total of 658 infants were analyzed and 196 (30%) had HiTG. One hundred thirty-six HiTG patients were matched with 136 CNTR. In the first 10 days of life, IV lipid, non-protein energy and total energy intakes, but not IV amino acids and carbohydrates, were significantly lower in HiTG infants. We found no differences between groups in diseases associated with prematurity. Anthropometry at 36 weeks (W), anthropometry at 2-year (Y) corrected age (CA), and neurodevelopment at 2Y CA were not different. Growth, diseases associated with prematurity, and neurodevelopment at 2Y CA in HiTG infants were similar to CNTR. This occurred despite a statistically significant albeit small reduction in IV lipid and non-protein energy intakes due to a strict TG monitoring and IV lipid titration at TG levels >250 mg/dL.

Does intravenous fish oil affect the growth of extremely low birth weight preterm infants on parenteral nutrition?

Long chain n-3 fatty acids (n-3 LCPUFA) play a pivotal role during central nervous system development and the provision of docosahexaenoic acid (DHA) is recommended for the preterm infant. However, there are concerns that oral fish oil, which is a good source of DHA, may adversely affect growth of preterm infants, as it decreases arachidonic acid (ARA). It has been about ten years since fish oil was added to the fat blend of intravenous (IV) lipid emulsions (LE) but information on growth and other clinical outcomes of preterm infants is still scarce. We studied the effect of fish oil containing IV LE vs standard IV LE on growth in a large cohort of preterm infants who received routine parenteral nutrition (PN). We retrospectively reviewed growth data of 546 preterm infants with a birth weight (BW)<1250g consecutively admitted to our NICU between Oct-2008 and Jun-2017 who received PN starting from the first day of life. Individual patients received only one of 5 commercially available IV LE. For the purpose of this study we grouped the patients who received the fish oil containing LE (IV-FO) and those who received conventional LE (CNTR). We compared PN and enteral nutrition (EN) intakes, and growth from birth to 36+0 weeks post-menstrual age (W PMA). Demographics, birth data and the incidence of the main complications of prematurity were similar between the two groups (IV-FO: n=240, Gestational age (GA) 197±16d, BW 942±181g; CNTR: n=237, GA 199±17d, BW 960±197g). No difference was found in PN and EN energy and macronutrient intakes from birth to 36+0W PMA, as well as in the proportion of human milk to infant milk formula. Weight gain from the regained BW to 36+0W PMA was slightly but significantly higher in IV-FO group: 17.3±2.8 and 16.8±2.7g∙kg-1∙d-1, IV-FO and CNTR respectively (p=0.03). There was no difference in length gain and head growth nor in body size at 36+0W PMA between the two groups. The use of IV fish oil did not negatively affect weight gain in a cohort of preterm infants. Large randomized controlled trials are needed to assess the effect of IV fish oil on the complication of prematurity and on selected domains of infant development.

Addition of probiotics to glutamine therapy as intestinal barrier protection for severe surgical patients

Objective: To study the protecting effect of combining probiotics with Glutamine (Gln) in intestinal barrier function of severe surgical patients. Methods: A total of 36 surgical patients with severe gastrointestinal diseases were randomly divided into routine Parenteral Nutrition (PN) group, probiotics group, and combination group, which received conventional PN, PN+probiotics and PN+probiotics+GLn respectively. Then all groups were measured the ratio of lactulose and Mannitol (L/M), the concentration of serum endotoxin, the level of intestinal Lactobacillus, and the number of gram-negative bacterium before operation and at the 3rd, 8th day after operation. Results: At 8th day, the ratio of lactulose and mannitol levels as well as the concentration of serum endotoxin in the probiotics group were much lower than those in the PN group (P<0.05), but much higher than the combination group (P<0.05)The level of intestinal Lactobacillus in the probiotics group was much higher than that in the PN group (P<0.05), but lower than the combination group (P<0.05). There was no significant difference in the number of gram-negative bacterium among the three groups (P>0.05). Conclusions: Addition of probiotics to Gln therapy has better protecting effect in intestinal barrier function.

Half-life of plasma phytosterols in very low birth weight preterm infants on routine parenteral nutrition with vegetable oil-based lipid emulsions

Phytosterols in vegetable oil (VO)-based lipid emulsions (LE) likely contribute to parenteral nutrition-associated cholestasis (PNAC) in preterm infants. No characterization of plasma phytosterol half-lives has been done in very low birth weight (VLBW) preterm infants receiving parenteral nutrition (PN) with LE. In a prospective cohort study, 45 VLBW preterm infants who received PN underwent serial blood sample measurements of sitosterol (SITO), campesterol (CAMP), and stigmasterol (STIGM). Plasma phytosterol half-lives were calculated from the phytosterol concentrations-decay curves by using a single-compartment model. After the stop of the intravenous LE, study infants had significantly lower plasma total CAMP, STIGM and SITO concentrations. The decay of plasma phytosterol concentrations was monoexponential. Half-life of plasma total CAMP, STIGM and SITO was 13.5±6.9, 10.3±4.5 and 10.3±4.0 days, respectively. Plasma phytosterol half-lives did not correlate with gestational age, birth weight, cumulative phytosterol intakes and plasma conjugated bilirubin. VLBW preterm infants on PN with LE had rather long plasma phytosterol half-lives similar to hypercholesterolemic adults and phytosterolemic homozygotes patients. We speculate that the accumulation of phytosterols could contribute to their vulnerability to PNAC. The Ethics Committee of Marche-Italy (DG/469); www.clinicaltrials.gov (identification number NCT02758834).

Phytosterol Esterification is Markedly Decreased in Preterm Infants Receiving Routine Parenteral Nutrition.

Several studies reported the association between total plasma phytosterol concentrations and the parenteral nutrition-associated cholestasis (PNAC). To date, no data are available on phytosterol esterification in animals and in humans during parenteral nutrition (PN). We measured free and esterified sterols (cholesterol, campesterol, stigmasterol, and sitosterol) plasma concentrations during PN in 16 preterm infants (500-1249g of birth weight; Preterm-PN), in 11 term infants (Term-PN) and in 12 adults (Adult-PN). Gas chromatography-mass spectrometry was used for measurements. Plasma concentrations of free cholesterol (Free-CHO), free phytosterols (Free-PHY) and esterified phytosterols (Ester-PHY) were not different among the three PN groups. Esterified cholesterol (Ester-CHO) was statistically lower in Preterm-PN than Adult-PN. Preterm-PN had significantly higher Free-CHO/Ester-CHO and Free-PHY/Ester-PHY ratios than Adult-PN (Free-CHO/Ester-CHO: 1.1±0.7 vs. 0.6±0.2; Free-PHY/Ester-PHY: 4.1±2.6 vs. 1.3±0.8; *P<0.05). Free-CHO/Ester-CHO and Free-PHY/Ester-PHY ratios of Term-PN (Free-CHO/Ester-CHO: 1.1±0.4; Free-PHY/Ester-PHY: 2.9±1.7) were not different from either Preterm-PN or from Adult-PN. Plasma Free-CHO/Ester-CHO and Free-PHY/Ester-PHY were unchanged after 24h on fat-free PN both in Preterm-PN and in Adult-PN. Free-PHY/Ester-PHY did not correlate with phytosterol intake in Preterm-PN. Free-PHY/Ester-PHY of Preterm-PN was positively correlated with the Free-CHO/Ester-CHO and negatively correlated with gestational age and birth weight. In conclusion, PHY were esterified to a lesser extent than CHO in all study groups; the esterification was markedly decreased in Preterm-PN compared to Adult-PN. The clinical consequences of these findings warrant further investigations.

Quality of newborn care: adherence to guidelines for parenteral nutrition in preterm infants in four European countries

ObjectiveThe level of adherence to guidelines should be explored particularly in preterm infants for whom poor nutrition has major effects on outcomes in later life. The objective was to evaluate...

The effect of 5 intravenous lipid emulsions on plasma phytosterols in preterm infants receiving parenteral nutrition: a randomized clinical trial

Elevated plasma phytosterol concentrations are an untoward effect of parenteral nutrition (PN) with vegetable oil-based lipid emulsions (LEs). Phytosterols are elevated in neonatal cholestasis, but the relation remains controversial. The objective was to study the effect of 5 LEs on plasma phytosterols in preterm infants. One hundred forty-four consecutive admitted preterm infants (birth weight: 500-1249 g) were studied. Patients were randomly assigned to receive 1 of 5 different LEs: S [100% soybean oil (SO)], MS [50% medium-chain triglycerides (MCTs) and 50% SO], MSF (50% MCTs, 40% SO, and 10% fish oil (FO)], OS (80% olive oil and 20% SO), or MOSF (30% MCTs, 25% olive oil, 30% SO, and 15% FO). Phytosterols in the LEs and in plasma (on postnatal day 7 and day 14) were measured by gas chromatography-mass spectrometry. Patients in the S group had significantly higher total phytosterol intakes than did the other study groups. On PN days 7 and 14, plasma phytosterol concentrations were highest in the S group and lowest in the MOSF group. Despite similar β-sitosterol intakes between the MS and MSF groups, plasma concentrations were significantly lower in the MSF than in the MS group. Only 3 patients (2.1%) developed cholestasis: 1 in the MS, 1 in the MSF, and 1 in the MOSF group. No cases of cholestasis were observed in the S and OS groups. In uncomplicated preterm infants receiving routine PN, we found a correlation between phytosterol intake and plasma phytosterol concentrations; however, cholestasis was rare and no difference in liver function at 6 wk was observed.

Effect of ω-3 polyunsaturated fatty acid on toll-like receptors in patients with severe multiple trauma

This study examined the effects of ω-3 polyunsaturated fatty acid (ω-3PUFA) on the expression of toll-like receptor 2 (TLR2), toll-like receptor 4 (TLR4) and some related inflammatory factors in peripheral blood mononuclear cells (PBMCs) of patients with early-stage severe multiple trauma. Thirty-two patients who were admitted to the Department of Traumatic Surgery, Tongji Hospital (Wuhan, China) between May 2010 and November 2010, and diagnosed as having severe multiple trauma with a injury severity score (ISS) no less than 16, were enrolled in the study and divided into two groups at random (n=16 in each): ω-3PUFA group and control group in which routine parenteral nutrition supplemented with ω-3PUFA or not was administered to the patients in two groups for consecutive 7 days. Peripheral blood from these patients was collected within 2 h of admission (day 0), and 1, 3, 5 and 7 days after the nutritional support. PBMCs were isolated and used for detection of the mRNA and protein expression of TLR2 and TLR4 by using real-time PCR and flow cytometry respectively, the levels of NF-κB by quantum dots-based immunofluorescence assay, the levels of TNF-α, IL-2, IL-6 and COX-2 by ELISA, respectively. The results showed that the mRNA and protein expression of TLR2 and TLR4 in PBMCs was significantly lower in ω-3PUFA group than in control group 5 and 7 days after nutrition support (both P<0.05). The levels of TNF-α, IL-2, IL-6 and COX-2 were found to be substantially decreased in PBMCs in ω-3PUFA group as compared with control group at 5th and 7th day (P<0.05 for all). It was concluded that ω-3PUFA can remarkably decrease the expression of TLR2, TLR4 and some related inflammatory factors in NF-κB signaling pathway in PBMCs of patients with severe multiple trauma, which suggests that ω-3PUFA may suppress the excessive inflammatory response meditated by the TLRs/NF-κB signaling pathway.

ESPEN Guidelines on Parenteral Nutrition: Non-surgical oncology

Parenteral nutrition offers the possibility of increasing or ensuring nutrient intake in patients in whom normal food intake is inadequate and enteral nutrition is not feasible, is contraindicated or is not accepted by the patient. These guidelines are intended to provide evidence-based recommendations for the use of parenteral nutrition in cancer patients. They were developed by an interdisciplinary expert group in accordance with accepted standards, are based on the most relevant publications of the last 30 years and share many of the conclusions of the ESPEN guidelines on enteral nutrition in oncology. Under-nutrition and cachexia occur frequently in cancer patients and are indicators of poor prognosis and, per se, responsible for excess morbidity and mortality. Many indications for parenteral nutrition parallel those for enteral nutrition (weight loss or reduction in food intake for more than 7-10 days), but only those who, for whatever reason cannot be fed orally or enterally, are candidates to receive parenteral nutrition. A standard nutritional regimen may be recommended for short-term parenteral nutrition, while in cachectic patients receiving intravenous feeding for several weeks a high fat-to-glucose ratio may be advised because these patients maintain a high capacity to metabolize fats. The limited nutritional response to the parenteral nutrition reflects more the presence of metabolic derangements which are characteristic of the cachexia syndrome (or merely the short duration of the nutritional support) rather than the inadequacy of the nutritional regimen. Perioperative parenteral nutrition is only recommended in malnourished patients if enteral nutrition is not feasible. In non-surgical well-nourished oncologic patients routine parenteral nutrition is not recommended because it has proved to offer no advantage and is associated with increased morbidity. A benefit, however, is reported in patients undergoing hematopoietic stem cell transplantation. Short-term parenteral nutrition is however commonly accepted in patients with acute gastrointestinal complications from chemotherapy and radiotherapy, and long-term (home) parenteral nutrition will sometimes be a life-saving maneuver in patients with sub acute/chronic radiation enteropathy. In incurable cancer patients home parenteral nutrition may be recommended in hypophagic/(sub)obstructed patients (if there is an acceptable performance status) if they are expected to die from starvation/under nutrition prior to tumor spread.

Why does the plasma urea concentration increase in acute dehydration?

Acute dehydration biochemically manifests itself with mild uremia and a normal creatinine concentration. The reason for this, as quoted in popular medical textbooks, is that there is increased reabsorption of urea by the kidneys ([1][1]). This explanation is rarely taken further by clinical tutors

Urinary pyridinium cross-link excretion is increased in critically ill surgical patients.

To determine: a) the rate of pyridinium cross-links of collagen excretion, breakdown products of bone, in critically ill surgical patients in the intensive care unit (ICU); and b) the relationship between cross-link excretion and nitrogen excretion and balance to ascertain whether collagen breakdown products contribute to protein losses during a hypercatabolic state. Observational study starting on the first postoperative day to 20 days or until discharge. A surgical ICU in a University hospital. Nine mechanically ventilated, postoperative surgical patients (73 +/- 3 [SD] yrs), receiving routine parenteral nutrition (18% protein) and 17 age-matched healthy subjects. None. Resting energy expenditure was determined daily for < or = 5 days after admission, and energy intake was set at 1.04 times the initial energy expenditure; thereafter, values of intake were reset weekly. Daily 24-hr urine samples were analyzed for cross-links, total and urea nitrogen, calcium, and creatinine for 20 days or until discharge. Two urine samples were also analyzed for cross-links in the healthy subjects. The excretion of cross-links from the surgical patients was markedly higher (p < .001) than in the healthy subjects, and calcium balance was significantly negative (p < .05). Patients who were discharged from the ICU within 5 days showed a lower rate of cross-link excretion (p < .02) and less day-to-day variability, compared with those patients who stayed longer, whether calculated over the course of the study or over the first 2 days in the ICU. There was no correlation between cross-links and energy expenditure, nitrogen excretion, or balance. The rate of cross-link excretion in critically ill patients: a) is markedly increased; b) is greater within the first two postoperative days in those patients who have an extended stay (> 5 days) in the ICU; and c) is independent of the rate of nitrogen excretion. These findings suggest that critically ill postoperative patients experience an acute breakdown of collagen, which is likely due to resorption of bone or possibly comes from other collagen sources.

Enteral nutrients promote postnatal maturation of intestinal motor activity in preterm infants.

Low-compliance perfusion manometry and clinical outcome were used to assess the chronic effects of feeding on functional maturation of the preterm intestine in 32 infants. During the first postnatal week, 16 infants received a small volume of formula to supplement their routine parenteral nutrition (24 ml.kg-1.day-1), and 16 received an equal volume of water. After 10 days of enteral "feedings," the manometric tracings of those infants who had received nutrient feedings had more migrating activity (P < 0.01) and less clustered phasic activity (P < 0.05) during fasting than did the tracings of those infants who had received nonnutrient feedings containing sterile water. Infants who received nutrient feedings demonstrated a change in motor activity in response to feeding; infants who had received water feedings failed to show a change in motor activity in response to feeding. After water-fed infants were given formula feedings for 2 wk, their motor activity patterns were similar to those who had initially been fed formula, and their motor responses to feeding were restored. Despite normalization of their motor activity patterns by 1 mo of age, these water-fed infants established full enteral nutrition and full nipple feedings later than did infants who had been fed formula (P < 0.01), suggesting that the delay of enteral feedings in preterm infants does not permit optimal intestinal maturation of nonmucosal functions.

Postnatal growth in infants born before 30 weeks' gestation.

The postnatal weight pattern up to 14 weeks after birth was determined in 184 singleton survivors born at 23 to 29 weeks' gestation in whom routine parenteral nutrition was used before milk feeding was established. A mean postnatal weight loss of 14% of birth weight occurred at a mean of 6 days. The more immature infants had significantly higher postnatal weight loss and longer time to regain birth weight despite a higher volume intake in the first week. From the fourth postnatal week all gestational subgroups had a mean weight gain at above intrauterine growth rate. As a result of the initial period of weight loss, however, the mean body weight remained below the 10th percentile of the intrauterine growth curve. The early growth rate in infants small for gestational age was higher than those who were appropriate weight for gestation, although the mean body weight of the former group remained significantly lower at 2 years.