Enteral nutrition and total parenteral nutrition components in the course of total parenteral nutrition–associated cholestasis in neonatal necrotizing enterocolitis Michelle Veenstra, MD,a Logan Danielson, BS,a Evan Brownie, MD,a May Saba, PharmD,c Girija Natarajan, MD,d and Michael Klein, MD,a,b Detroit, MI
Background. Newborns with necrotizing enterocolitis (NEC) are at high risk for the development of total parenteral nutritional–associated cholestasis (TPNAC). Patients with NEC were evaluated to determine risk factors for development of TPNAC and predictors of resolution. We hypothesized that there are additional factors relating to the timing of enteral nutrition or TPN components that effect development and persistence of TPNAC in patients with NEC that may be altered to decrease the chance of progression to liver failure. Methods. This was a retrospective chart review of NEC patients from 2001 to 2010. TPNAC was defined as direct bilirubin $2 mg/dL, the level used for cholestasis in neonatal studies relating to TPNAC. Results. Of 178 patients with NEC, 96 developed TPNAC, and in 27 TPNAC had resolved by discharge. Days of TPN did not affect TPNAC resolution by discharge (P = 1.0). TPNAC was less likely to occur in patients with body weights >1,500 g, enteral nutrition within the first week of life, no infection, fewer TPN days, and lesser hospital stay (P < .01). There were many factors affecting resolution of cholestasis. Enteral nutrition within 6 days of birth decreased development of TPNAC (P < .01), and resumption of enteral nutrition within 3 weeks after NEC diagnosis increased the resolution of cholestasis (P < .01). No component of TPN was important for the development or resolution of cholestasis. Conclusion. Of the factors that effect development and resolution of TPNAC in NEC, the ones that we can alter include early enteral feeds and surveillance for infection. (Surgery 2014;156:578-83.) From the Department of Pediatric Surgery,a Children’s Hospital of Michigan, Department of Surgery,b Wayne State University School of Medicine, and Department of Pharmacyc and Division of Neonatal and Perinatal Medicine,d Children’s Hospital of Michigan, Detroit, MI
TOTAL PARENTERAL NUTRITION (TPN) often is required for the management of necrotizing enterocolitis (NEC) to prevent dehydration and malnutrition while enteral nutrition (EN) is withheld. While on TPN, patients are at risk of developing TPN-associated cholestasis (TPNAC), a
Presented at the 9th Annual Academic Surgical Congress in San Diego, CA, February 4–6, 2014. Accepted for publication April 15, 2014. Reprint requests: Michelle Veenstra, MD, Department of Pediatric Surgery, Children’s Hospital of Michigan, 3901 Beaubien Blvd, Detroit, MI 48201. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2014 Mosby, Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2014.04.031
578 SURGERY
precursor to liver failure and associated complications that cause significant morbidity.1 TPNAC is associated with a high rate of mortality among children with intestinal failure.2 Patients with NEC are at a greater risk for development of TPNAC than other neonates.1,3,4 Previous studies evaluating risk factors for TPNAC include patient populations of all neonates or all surgical neonates. NEC has a very different pathology than other operative conditions leading to TPN dependence such as atresia, gastroschisis, and other anomalies. We hypothesize that there are additional factors that effect development and persistence of TPNAC in patients with NEC pathology. These factors could include timing of EN before and after NEC diagnosis, presence of infection, or the components of TPN. Identification of
Surgery Volume 156, Number 3
these factors may assist with resolution of cholestasis before the development of parenteral nutrition–associated liver disease in this patient population. This study determined risk factors for development of TPNAC and resolution of cholestasis in NEC to remove confounding variables from evaluating patients with multiple pathologies. MATERIALS AND METHODS All patients admitted to the Neonatal Intensive Care Unit at The Children’s Hospital of Michigan with Bell’s Stage II and III NEC5 from 2001 to 2010 were included. All patients were outborn. After we obtained approval from the institutional research board, a retrospective chart review was conducted. Patients were excluded if they were missing direct bilirubin data. Functional cholestasis was defined as a direct bilirubin $2 mg/dL, the diagnostic criteria for TPNAC.6-8 At this point, bile flow is decreased, the ability to secrete bile acids and bilirubin is less, and within 7–29 days the hepatic architecture is distressed and bile stasis can be seen histologically.6 Patients were placed into three categories: no TPNAC, TPNAC that resolved, and TPNAC that did not resolve by discharge or death. Factors relating to patient demographics, NEC diagnosis, treatment, TPN characteristics, and operative details were recorded. An infection was defined as a new positive blood culture. If longterm TPN was expected or cholestasis identified, intravenous fat emulsion restriction was used (1 g/ kg three times per week).1 EN was defined as any amount of oral feeds and patients were started and maintained on trophic feeds whenever possible. Data were analyzed by a statistician using SPSS (IBM Corporation 2011; IBM Corporation, Somers, NY). Continuous variables were analyzed with analysis of variance and Bonferroni significance. Other variables were analyzed with likelihood ratio asymptotic significance (2-sided) and linear-bylinear asymptotic significance (2-sided). Relative risk (RR) was calculated for developing TPNAC and non-resolution of TPNAC by discharge. A linear regression model for gestational age and days on TPN was completed with t-statistics and two-tailed P values. RESULTS Patient demographics are listed in Table I. Of 178 patients included, 82 patients (46%) did not develop TPNAC and of the remaining 96 patients who developed TPNAC, 27 (28% of TPNAC
Veenstra et al 579
patients) resolved cholestasis before discharge. Mortality was 33%; 31 did not develop TPNAC, 28 developed TPNAC, 3 that resolved, and 25 that did not resolve. Small for gestational age (SGA) accounted for 8% (14 patients) and did not effect TPNAC development (P = .09) but did make a patient 98.5% more likely to experience persisting TPNAC (P = .03; Tables II and III). Table II lists factors associated with development of TPNAC in patients with NEC. The number of days on TPN predicted TPNAC diagnosis (P < .01; Table II) but not resolution (P = 1). Patients who did not develop TPNAC averaged 24 days on TPN, patients with TPNAC resolution 81 days, and nonresolution 84 days. Patients on TPN at the time of NEC diagnosis were more likely to develop TPNAC and be cholestatic at discharge (Table III). In evaluating components of TPN, we found the amount of protein in development of TPNAC or resolution, the percent of calories of lipids and dextrose between TPNAC groups, and the amount of lipids infused with TPN at the time of NEC diagnosis or TPNAC diagnosis were not significant (P > .1 all). The dextrose concentration of TPN at peak direct bilirubin was indicative of the development of TPNAC (P < .01) but not resolution (P = .06). When a patient remained on TPN at discharge, they were 58% more likely to develop TPNAC and 55% more likely to remain cholestatic (RR = 1.6). The length of nil per os (NPO) at birth was important. EN within the first week of life decreased the chance of developing TPNAC (Table III; Fig). Receiving EN at least once made one only 61% as likely to develop TPNAC and only 43% as likely to have TPNAC at discharge (RR = 0.6–0.4). Receiving EN before being diagnosed with NEC decreased the likelihood of developing TPNAC and remaining cholestatic at discharge (Table III). Reaching one third or full EN before discharge did not increase the chance for resolution of TPNAC (P = .45, P = .28; Table III). The duration of NPO after the diagnosis of NEC correlated with TPNAC course (P < .01). When the patient’s first feed was after the NEC course, TPNAC developed in 74% of patients, and 59% remained cholestatic at discharge. A patient with less then 3 weeks of NPO had the least risk of TPNAC persisting to discharge. Patients with EN before NEC but not after developed TPNAC at a rate near that of the patients with
Background. Newborns with necrotizing enterocolitis (NEC) are at high risk for the development of total parenteral nutritional–associated cholestasis (TPNAC). Patients with NEC were evaluated to determine risk factors for development of TPNAC and predictors of resolution. We hypothesized that there are additional factors relating to the timing of enteral nutrition or TPN components that effect development and persistence of TPNAC in patients with NEC that may be altered to decrease the chance of progression to liver failure. Methods. This was a retrospective chart review of NEC patients from 2001 to 2010. TPNAC was defined as direct bilirubin $2 mg/dL, the level used for cholestasis in neonatal studies relating to TPNAC. Results. Of 178 patients with NEC, 96 developed TPNAC, and in 27 TPNAC had resolved by discharge. Days of TPN did not affect TPNAC resolution by discharge (P = 1.0). TPNAC was less likely to occur in patients with body weights >1,500 g, enteral nutrition within the first week of life, no infection, fewer TPN days, and lesser hospital stay (P < .01). There were many factors affecting resolution of cholestasis. Enteral nutrition within 6 days of birth decreased development of TPNAC (P < .01), and resumption of enteral nutrition within 3 weeks after NEC diagnosis increased the resolution of cholestasis (P < .01). No component of TPN was important for the development or resolution of cholestasis. Conclusion. Of the factors that effect development and resolution of TPNAC in NEC, the ones that we can alter include early enteral feeds and surveillance for infection. (Surgery 2014;156:578-83.) From the Department of Pediatric Surgery,a Children’s Hospital of Michigan, Department of Surgery,b Wayne State University School of Medicine, and Department of Pharmacyc and Division of Neonatal and Perinatal Medicine,d Children’s Hospital of Michigan, Detroit, MI
TOTAL PARENTERAL NUTRITION (TPN) often is required for the management of necrotizing enterocolitis (NEC) to prevent dehydration and malnutrition while enteral nutrition (EN) is withheld. While on TPN, patients are at risk of developing TPN-associated cholestasis (TPNAC), a
Presented at the 9th Annual Academic Surgical Congress in San Diego, CA, February 4–6, 2014. Accepted for publication April 15, 2014. Reprint requests: Michelle Veenstra, MD, Department of Pediatric Surgery, Children’s Hospital of Michigan, 3901 Beaubien Blvd, Detroit, MI 48201. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2014 Mosby, Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2014.04.031
578 SURGERY
precursor to liver failure and associated complications that cause significant morbidity.1 TPNAC is associated with a high rate of mortality among children with intestinal failure.2 Patients with NEC are at a greater risk for development of TPNAC than other neonates.1,3,4 Previous studies evaluating risk factors for TPNAC include patient populations of all neonates or all surgical neonates. NEC has a very different pathology than other operative conditions leading to TPN dependence such as atresia, gastroschisis, and other anomalies. We hypothesize that there are additional factors that effect development and persistence of TPNAC in patients with NEC pathology. These factors could include timing of EN before and after NEC diagnosis, presence of infection, or the components of TPN. Identification of
Surgery Volume 156, Number 3
these factors may assist with resolution of cholestasis before the development of parenteral nutrition–associated liver disease in this patient population. This study determined risk factors for development of TPNAC and resolution of cholestasis in NEC to remove confounding variables from evaluating patients with multiple pathologies. MATERIALS AND METHODS All patients admitted to the Neonatal Intensive Care Unit at The Children’s Hospital of Michigan with Bell’s Stage II and III NEC5 from 2001 to 2010 were included. All patients were outborn. After we obtained approval from the institutional research board, a retrospective chart review was conducted. Patients were excluded if they were missing direct bilirubin data. Functional cholestasis was defined as a direct bilirubin $2 mg/dL, the diagnostic criteria for TPNAC.6-8 At this point, bile flow is decreased, the ability to secrete bile acids and bilirubin is less, and within 7–29 days the hepatic architecture is distressed and bile stasis can be seen histologically.6 Patients were placed into three categories: no TPNAC, TPNAC that resolved, and TPNAC that did not resolve by discharge or death. Factors relating to patient demographics, NEC diagnosis, treatment, TPN characteristics, and operative details were recorded. An infection was defined as a new positive blood culture. If longterm TPN was expected or cholestasis identified, intravenous fat emulsion restriction was used (1 g/ kg three times per week).1 EN was defined as any amount of oral feeds and patients were started and maintained on trophic feeds whenever possible. Data were analyzed by a statistician using SPSS (IBM Corporation 2011; IBM Corporation, Somers, NY). Continuous variables were analyzed with analysis of variance and Bonferroni significance. Other variables were analyzed with likelihood ratio asymptotic significance (2-sided) and linear-bylinear asymptotic significance (2-sided). Relative risk (RR) was calculated for developing TPNAC and non-resolution of TPNAC by discharge. A linear regression model for gestational age and days on TPN was completed with t-statistics and two-tailed P values. RESULTS Patient demographics are listed in Table I. Of 178 patients included, 82 patients (46%) did not develop TPNAC and of the remaining 96 patients who developed TPNAC, 27 (28% of TPNAC
Veenstra et al 579
patients) resolved cholestasis before discharge. Mortality was 33%; 31 did not develop TPNAC, 28 developed TPNAC, 3 that resolved, and 25 that did not resolve. Small for gestational age (SGA) accounted for 8% (14 patients) and did not effect TPNAC development (P = .09) but did make a patient 98.5% more likely to experience persisting TPNAC (P = .03; Tables II and III). Table II lists factors associated with development of TPNAC in patients with NEC. The number of days on TPN predicted TPNAC diagnosis (P < .01; Table II) but not resolution (P = 1). Patients who did not develop TPNAC averaged 24 days on TPN, patients with TPNAC resolution 81 days, and nonresolution 84 days. Patients on TPN at the time of NEC diagnosis were more likely to develop TPNAC and be cholestatic at discharge (Table III). In evaluating components of TPN, we found the amount of protein in development of TPNAC or resolution, the percent of calories of lipids and dextrose between TPNAC groups, and the amount of lipids infused with TPN at the time of NEC diagnosis or TPNAC diagnosis were not significant (P > .1 all). The dextrose concentration of TPN at peak direct bilirubin was indicative of the development of TPNAC (P < .01) but not resolution (P = .06). When a patient remained on TPN at discharge, they were 58% more likely to develop TPNAC and 55% more likely to remain cholestatic (RR = 1.6). The length of nil per os (NPO) at birth was important. EN within the first week of life decreased the chance of developing TPNAC (Table III; Fig). Receiving EN at least once made one only 61% as likely to develop TPNAC and only 43% as likely to have TPNAC at discharge (RR = 0.6–0.4). Receiving EN before being diagnosed with NEC decreased the likelihood of developing TPNAC and remaining cholestatic at discharge (Table III). Reaching one third or full EN before discharge did not increase the chance for resolution of TPNAC (P = .45, P = .28; Table III). The duration of NPO after the diagnosis of NEC correlated with TPNAC course (P < .01). When the patient’s first feed was after the NEC course, TPNAC developed in 74% of patients, and 59% remained cholestatic at discharge. A patient with less then 3 weeks of NPO had the least risk of TPNAC persisting to discharge. Patients with EN before NEC but not after developed TPNAC at a rate near that of the patients with
