Early cerebral and intestinal oxygenation in the risk assessment of necrotizing enterocolitis in preterm infants.

BACKGROUND AND AIM: Predicting necrotizing enterocolitis (NEC) might help in preventing its devastating consequences. We aimed to investigate whether early cerebral and intestinal tissue oxygen saturation (rSO2) and fractional tissue oxygen extraction (FTOE) predict the onset of NEC. STUDY DESIGN: Prospective observational case-control study. SUBJECTS: Infants with gestational age (GA) <32 weeks were included. For every NEC case we matched two controls based on GA, birth weight (BW), and a patent ductus arteriosus. OUTCOME MEASURES: Cerebral oxygenation and intestinal oxygenation were prospectively monitored two-hours daily during the first five days after birth and once a week thereafter until five weeks after birth or until NEC developed. We used Kaplan-Meier analyses to determine the ability of near-infrared spectroscopy (NIRS) measurements, including their variability, to predict the development of NEC. RESULTS: We included ten infants (median (range) GA 27.1 (24.6-29.4) weeks, BW 903 (560-1630) grams) who developed NEC at median postnatal day 13 (range: 4-43 days), and 20 matched controls. Infants with cerebral rSO2 <70% within the first 48 h after birth developed NEC significantly more often than infants with cerebral rSO2 ≥70% (odds ratio 9.00 (95% CI 1.33-61.14). Intestinal FTOE was higher in infants who developed NEC compared to controls during the last NIRS measurement at median 2 days (range: 1-7) before NEC onset (median 0.65 vs. 0.44). CONCLUSIONS: Cerebral oxygenation monitoring early after birth might be valuable in the risk assessment of NEC development. Additionally, our results suggest that intestinal oxygenation is impaired before the onset of clinical NEC.

Fecal Bile Salts and the Development of Necrotizing Enterocolitis in Preterm Infants.

BACKGROUND: Intestinal bile salts (BSs) may be implicated in NEC development. We hypothesized that fecal BS levels are higher in preterm infants at risk for NEC. METHODS: We compared the composition and concentration of fecal BSs in ten preterm infants who developed NEC (Bell's Stage ≥ II) with twenty matched control infants without NEC. Conjugated and unconjugated fecal BSs were measured after birth (T1) and twice prior to NEC (T2, T3). Data are presented as medians and interquartile ranges. RESULTS: GA and BW were similar in all preterms: ~27+4 weeks and ~1010 g. Age of NEC onset was day 10 (8-24). T1 was collected 2 (1-3) days after birth. T2 and T3 were collected 5 (5-6) days and 1 (0-2) day before NEC or at corresponding postnatal ages in controls. The composition of conjugated BSs did not differ between the two groups. Total unconjugated BSs were 3-fold higher before NEC compared to controls at corresponding ages (0.41 µmol/g feces (0.21-0.74) versus 0.14 µmol/g feces (0.06-0.46), p < 0.05). CONCLUSION: Fecal BS concentrations are higher in preterm infants who develop NEC compared to infants without NEC. Further study is needed to determine the predictive value of fecal BSs in the development of NEC.

Multisite Tissue Oxygenation Monitoring Indicates Organ-Specific Flow Distribution and Oxygen Delivery Related to Low Cardiac Output in Preterm Infants With Clinical Sepsis.

OBJECTIVES: Cardiac output may be compromised in preterm infants with sepsis. Whether low cardiac output is associated with low tissue oxygen supply in these patients is unclear. The aim of the current study was to assess the association between cardiac output, assessed by echocardiography, and tissue oxygenation, measured with multisite near-infrared spectroscopy, in a cohort of preterm infants with clinical sepsis. DESIGN: Prospective observational cohort study. SETTING: Level III neonatal ICU. PATIENTS: Twenty-four preterm infants (gestational age < 32 wk) with clinical sepsis. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Clinical and echocardiographic assessment of hemodynamics was performed within 48 hours of sepsis workup and repeated at least 24 hours later. We measured cerebral, renal, and intestinal tissue oxygen saturation using near-infrared spectroscopy during an hour of stable measurements directly preceding or following echocardiography and calculated fractional tissue oxygen extraction in each tissue. We determined Spearman correlation coefficients between fractional tissue oxygen extraction and right ventricular output corrected for patent foramen ovale flow, left ventricular output corrected for ductus arteriosus flow, and superior vena cava flow. Right ventricular output corrected for patent foramen ovale and left ventricular output corrected for ductus arteriosus flow both correlated significantly with intestinal fractional tissue oxygen extraction (ρ, -0.700; p = 0.036 and ρ, -0.604; p = 0.029, respectively). In contrast, no significant correlations were found between cardiac output measurements and cerebral and renal fractional tissue oxygen extraction, respectively. Changes in cardiac output measurements were not associated with observed changes in fractional tissue oxygen extraction values. CONCLUSIONS: Right ventricular output corrected for patent foramen ovale and left ventricular output corrected for ductus arteriosus flow, indicators of systemic blood flow in preterm infants with shunts, were negatively associated with intestinal fractional tissue oxygen extraction, but not with renal and cerebral fractional tissue oxygen extraction. These findings suggest that during low output states due to clinical sepsis intestinal perfusion is most at risk.

Near-Infrared Spectroscopy to Predict the Course of Necrotizing Enterocolitis.

OBJECTIVES: To investigate whether cerebral, liver, and infraumbilical regional tissue oxygen saturation (rSO2) and fractional tissue oxygen extraction (FTOE) could be used to diagnose necrotizing enterocolitis (NEC) and complicated NEC (Bell's stage 3B or death) during its early stages. METHODS: A prospective observational cohort study of preterm infants with suspected or diagnosed NEC. We compared the mean eight-hour cerebral, liver, and infraumbilical rSO2 and FTOE values of infants with no NEC and definite NEC and of infants with uncomplicated and complicated NEC in the first forty-eight hours after onset of symptoms, suspicious for NEC. Furthermore, we determined cut-off values by generating receiver operating characteristics curves in case of significant differences in the first eight-hour mean values of rSO2 between infants with no NEC and definite NEC and between infants with uncomplicated and complicated NEC. RESULTS: We included 33 patients: 13 no NEC, 10 with uncomplicated NEC, and 10 with complicated NEC. We found no significant differences in the first twenty-four hours after onset of symptoms in rSO2 and FTOE between infants with no NEC and definite NEC. In preterm infants with complicated NEC, we observed significantly lower cerebral, liver, and infraumbilical rSO2 and higher FTOE within twenty-four hours after onset of symptoms compared with infants with uncomplicated NEC. A continuous cerebral rSO2 ≤ 71% and liver rSO2 ≤ 59% in the first eight hours after onset of symptoms predicted the onset of complicated NEC with a sensitivity of 1.0 and specificity of 0.8, and a sensitivity of 1.0 and specificity of 1.0, respectively. CONCLUSIONS: By measuring the cerebral and splanchnic oxygenation it is possible to differentiate complicated NEC from uncomplicated NEC. In our sample, NIRS monitoring did not proof useful for distinguishing between definite NEC and no NEC in preterm infants with clinical signs suspicious of NEC.

The relation between splanchnic ischaemia and intestinal damage in necrotising enterocolitis.

OBJECTIVES: The underlying pathophysiology of necrotising enterocolitis (NEC) remains incompletely understood, particularly the role of intestinal perfusion. We aimed to determine the relation between cerebral and splanchnic fractional tissue oxygen extraction (FTOE), a marker for tissue underperfusion, with intestinal fatty acid-binding protein in plasma (I-FABPp), a marker for intestinal damage, in infants with NEC. Furthermore, we investigated the combined courses of cerebral and splanchnic FTOE values and I-FABPp levels in uncomplicated (conservative treatment) and complicated NEC (surgery or death). DESIGN: This study was part of a prospective observational cohort study. PATIENTS: We included 19 preterm infants with NEC (9 uncomplicated, 10 complicated). INTERVENTIONS: Using near-infrared spectroscopy, we measured regional cerebral and splanchnic tissue oxygen saturations continuously for 48 h after NEC onset. We measured I-FABPp levels simultaneously. MAIN OUTCOME MEASURES: We used Spearman correlation tests to calculate correlation coefficients between FTOE values and I-FABPp levels in uncomplicated and complicated NEC. RESULTS: Median (range) gestational age was 28 (25-36) weeks and median (range) birth weight was 1290 (740-2400) g. Cerebral and splanchnic FTOE values correlated strongly with I-FABPp levels (rho between .745 and 0.900; p<0.001-0.037) during the first 16 h after NEC onset. Thereafter, in uncomplicated NEC, splanchnic FTOE values increased while I-FABPp levels decreased concomitantly. In complicated NEC both splanchnic FTOE values and I-FABPp levels decreased. CONCLUSIONS: Combining cerebral and splanchnic FTOE values with I-FABPp levels, gives insight in the pathological chain of events resulting in progression or recovery of intestinal ischaemia in NEC. TRIAL REGISTRATION NUMBER: NTR3239.

The Effect of Maternal Antihypertensive Drugs on the Cerebral, Renal and Splanchnic Tissue Oxygen Extraction of Preterm Neonates.

BACKGROUND: Drugs with antihypertensive action are frequently used in obstetrics for the treatment of preeclampsia (labetalol) and tocolysis (nifedipine) or for neuroprotection (MgSO4), and may affect the hemodynamics of preterm born neonates. OBJECTIVE: The aim of this study was to assess whether maternal antihypertensive drugs affect multisite oxygenation levels of the neonate. METHODS: Eighty preterm neonates of ≤32 weeks of gestational age were monitored using near-infrared spectroscopy. Mean cerebral, renal and splanchnic fractional tissue oxygen extractions (cFTOE, rFTOE and sFTOE) were calculated for the first 5 postnatal days. We determined the effect of various maternal antihypertensive drugs on cFTOE and rFTOE using multilevel analysis, and on sFTOE using Kruskal-Wallis and Mann-Whitney U tests. RESULTS: Eleven infants were exposed to labetalol ± MgSO4, 7 to nifedipine ± MgSO4, 20 to MgSO4 only, and 42 to no maternal antihypertensive drugs. The infants exposed to labetalol ± MgSO4 had a lower cFTOE on days 1 (0.14, p = 0.031), 2 (0.13, p = 0.035) and 4 (0.18, p = 0.046) than nonexposed infants on the corresponding days (0.22, 0.20 and 0.24, respectively). On day 2, cFTOE was also lower in infants exposed to nifedipine ± MgSO4 (0.11, p = 0.028) and to MgSO4 only (0.15, p = 0.047). sFTOE was higher in infants exposed to labetalol ± MgSO4 on days 1 (µ = 0.71) and 2 (µ = 0.82) than in nonexposed infants (µ = 0.26, p = 0.04 and µ = 0.55, p = 0.007, respectively). Maternal antihypertensive drugs did not affect rFTOE. CONCLUSIONS: Low neonatal cFTOE found with maternal antihypertensive drug exposure may relate to either increased cerebral perfusion or neurologic depression induced by the medication, or preferential brain perfusion associated with preeclampsia placental insufficiency. Concomitantly high sFTOE found with labetalol exposure supports the latter, while renal autoregulation may explain rFTOE stability.

A Hemodynamically Significant Patent Ductus Arteriosus Does Not Affect Cerebral or Renal Tissue Oxygenation in Preterm Infants.

BACKGROUND: Patent ductus arteriosus (PDA) is common in preterm infants and is associated with significant morbidity. To determine whether the PDA is hemodynamically significant (HSDA), several echocardiographic parameters have been suggested, including retrograde diastolic blood flow in the descending aorta (Dao). OBJECTIVE: To assess the impact of an HSDA, including retrograde diastolic flow in the Dao, on regional tissue oxygen saturation (rSO2) and extraction measured by near-infrared spectroscopy (NIRS). METHODS: This is a prospective observational cohort study in which we included preterm infants (GA <32 weeks) who underwent echocardiographic screening because of clinical signs of an HSDA within 2 weeks after birth. We measured cerebral and renal rSO2 on the day of echocardiography. HSDA was diagnosed if left-to-right shunting through the PDA was accompanied by left atrial-to-aortic root ratio >1.4 and/or left pulmonary artery end-diastolic flow velocity >0.2 m/s and/or retrograde diastolic blood flow in the Dao. RESULTS: Forty-nine infants were included, with a median GA of 27.6 weeks (IQR: 26.1-29.0), birth weight of 980 g (IQR: 800-1,200), and postnatal age of 77 h (IQR: 70-107). Infants with a closed duct (n = 11), a non-HSDA (n = 18), and an HSDA (n = 20) had similar cerebral and renal NIRS measurements. Retrograde diastolic blood flow in the Dao, present in 11 infants with PDA, also did not affect cerebral and renal NIRS measurements. CONCLUSION: In preterm infants with clinical signs of an HSDA within 2 weeks after birth, cerebral and renal oxygen saturation and extraction are not affected by an HSDA or by retrograde diastolic blood flow in the Dao.

Assessing cerebrovascular autoregulation in infants with necrotizing enterocolitis using near-infrared spectroscopy.

BACKGROUND: We assessed cerebrovascular autoregulation (CAR) in preterm infants with definite necrotizing enterocolitis (NEC), Bell's stage 2 or 3, and infants without NEC, using near-infrared spectroscopy. We hypothesized that CAR would be more often impaired in infants with NEC compared with infants without NEC. METHODS: We measured cerebral regional tissue oxygen saturation, arterial oxygen saturation, and mean arterial blood pressure (MABP) during 48 h. We calculated the correlation between cerebral fractional tissue oxygen extraction and MABP for each patient. A statistically significant negative correlation reflected impaired CAR. RESULTS: We included 15 infants with definite NEC (median (range) gestational age 27.4 (25.6-34.7) wk; birth weight 1,070 (670-2,400) g) and 13 infants without NEC (gestational age 27.9 (26.3-34.7) wk; birth weight 980 (640-2,640) g). Fourteen infants had a statistically significant negative correlation (ρ -0.468 to-0.104), of whom five were infants without NEC (5/13; 38%) and nine with definite NEC (9/15; 60%). The difference in prevalence of impaired CAR was not statistically significant. CONCLUSION: Impaired CAR is present in a substantial proportion of infants with definite NEC, which may predispose them to NEC-associated neurological damage.

The Association between Multisite Near-Infrared Spectroscopy and Routine Hemodynamic Measurements in Relation to Short-Term Outcome in Preterms with Clinical Sepsis.

BACKGROUND: The added clinical value of multisite near-infrared spectroscopy (NIRS) monitoring to detect low organ tissue perfusion in preterm infants at risk of circulatory failure remains unclear. OBJECTIVES: To evaluate the associations between multisite NIRS measurements and clinical signs of circulatory failure in relation to short-term outcome in preterm infants with clinical sepsis. METHODS: Prospective cohort study of preterm infants (gestational age <32 weeks) with clinical sepsis. We monitored cerebral, renal, and intestinal oxygen saturation using NIRS for 72 h following sepsis workup and calculated fractional tissue oxygen extraction (FTOE). We recorded clinical signs of circulatory failure every 8 h. We analyzed the associations between FTOE values, clinical signs of circulatory failure, and short-term outcome. RESULTS: In 28 preterm infants with clinical sepsis, intraindividual and interindividual associations between NIRS values and clinical signs of circulatory failure were weak. At several points of time during the study period, cerebral and renal FTOE were higher in infants who developed intestinal complications compared with infants who did not, while clinical signs of circulatory failure never differed between groups. After correcting for multiple testing, significant differences disappeared. CONCLUSIONS: The associations between multisite FTOE values and clinical signs of circulatory failure were weak in preterm infants with clinical sepsis. Nevertheless, in contrast to clinical signs of circulatory failure, cerebral and renal FTOE values were associated with adverse short-term intestinal outcome in the uncorrected analyses. Multisite NIRS monitoring might help to detect critically low tissue oxygen delivery leading to adverse intestinal outcome not detected by routine hemodynamic measurements.

Abdominal near-infrared spectroscopy in preterm infants: a comparison of splanchnic oxygen saturation measurements at two abdominal locations.

BACKGROUND: Splanchnic tissue oxygenation monitoring has been performed at both the liver and the infra-umbilical regions. It is unknown whether these measurements could be substituted one for the other when interpreting splanchnic oxygenation since they have not been measured simultaneously before. AIMS: To evaluate the feasibility and safety of liver and infra-umbilical near-infrared spectroscopy (NIRS) monitoring in preterm infants with suspected necrotizing enterocolitis (NEC) and to assess the correlation and agreement between NIRS measurements performed simultaneously at the two abdominal locations. STUDY DESIGN AND SUBJECTS: This study was part of a prospective observational cohort study. Preterm infants who were suspected of NEC or who had been diagnosed with NEC were included. OUTCOME MEASURES: Liver oxygen saturation and infra-umbilical oxygen saturation were monitored simultaneously and continuously for 48h by NIRS. RESULTS: NIRS monitoring was performed in 20 out of 24 infants for the entire 48-hour study period. No adverse effects were observed. Values of liver and infra-umbilical oxygen saturation correlated weakly (Spearman's rho=0.244, P<.001). On the Bland-Altman plot liver oxygen saturation was higher than infra-umbilical oxygen saturation (mean difference 6.6%, SD 22.5%). CONCLUSIONS: Using NIRS as method for monitoring oxygen saturation simultaneously in both the liver and infra-umbilical regions is safe and feasible. Additionally, we demonstrated that values of liver and infra-umbilical oxygen saturation cannot be randomly substituted one for the other for the purpose of assessing splanchnic oxygenation.