Neuroapoptosis in newborns with respiratory acidosis at birth.

BACKGROUND: S100B protein is one of the most accurate biomarkers for diagnosis of neuroapoptosis and brain damage. The aim was to evaluate the lactate concentration and acid-base balance (pH, pCO2, pO2, HCO3c and BEb) in umbilical cord blood to predict high risk of neuroapoptosis and analyze the relationship between the levels of these biomarkers and umbilical cord blood S100B protein concentration at birth. METHODS: Apparently healthy newborns were included. S100B protein and blood gas test (lactate and acid-base balance) were determined in umbilical cord blood at birth. Newborns were classified into two groups: with and without high risk of neuroapoptosis. Newborns with high umbilical cord blood S100B protein concentration were considered newborns at high risk of neuroapoptosis. RESULTS: Sixty-one newborns were included, 12 had high risk of neuroapoptosis and 49 did not. S100B protein concentration correlate directly with pCO2 levels (Rho: 0.286, p = .0321) and lactate concentration (Rho: 0.278, p = .0315); and indirectly with pH (Rho: -0.332, p = .01). The analysis of the ROC curves yielded significant curves for pH and pCO2 to predict high risk of neuroapoptosis, pH optimal cutoff value was 7.19 (sensitivity: 50%, specificity: 83.7%, AUC: 0.708); and pCO2 optimal cutoff value was 60 mmHg (sensitivity: 30%, specificity: 85.4%, AUC: 0.705). CONCLUSIONS: Respiratory acidosis is associated to high concentrations of S100B protein in umbilical cord blood at birth. Umbilical cord blood pH and pCO2 may be useful in differentiating newborns at high risk of neuroapoptosis. Umbilical cord blood gas test may be valuable as risk indicator for neuroapoptosis at birth.

The choroid plexus sodium-bicarbonate cotransporter NBCe2 regulates mouse cerebrospinal fluid pH.

KEY POINTS: Normal pH is crucial for proper functioning of the brain, and disorders increasing the level of CO2 in the blood lead to a decrease in brain pH. CO2 can easily cross the barriers of the brain and will activate chemoreceptors leading to an increased exhalation of CO2 . The low pH, however, is harmful and bases such as HCO3- are imported across the brain barriers in order to normalize brain pH. We show that the HCO3- transporter NBCe2 in the choroid plexus of the blood-cerebrospinal fluid barrier is absolutely necessary for normalizing CSF pH during high levels of CO2 . This discovery represents a significant step in understanding the molecular mechanisms behind regulation of CSF pH during acid-base disturbances, such as chronic lung disease. ABSTRACT: The choroid plexus epithelium (CPE) is located in the brain ventricles where it produces the majority of the cerebrospinal fluid (CSF). The hypothesis that normal brain function is sustained by CPE-mediated CSF pH regulation by extrusion of acid-base equivalents was tested by determining the contribution of the electrogenic Na+ -HCO3- cotransporter NBCe2 to CSF pH regulation. A novel strain of NBCe2 (Slc4a5) knockout (KO) mice was generated and validated. The base extrusion rate after intracellular alkalization was reduced by 77% in NBCe2 KO mouse CPE cells compared to control mice. NBCe2 KO mice and mice with CPE-targeted NBCe2 siRNA knockdown displayed a reduction in CSF pH recovery during hypercapnia-induced acidosis of approximately 85% and 90%, respectively, compared to control mice. NBCe2 KO did not affect baseline respiration rate or tidal volume, and the NBCe2 KO and wild-type (WT) mice displayed similar ventilatory responses to 5% CO2 exposure. NBCe2 KO mice were not protected against pharmacological or heating-induced seizure development. In conclusion, we establish the concept that the CPE is involved in the regulation of CSF pH by demonstrating that NBCe2 is necessary for proper CSF pH recovery after hypercapnia-induced acidosis.

Distinct mechanisms underlie adaptation of proximal tubule Na+/H+ exchanger isoform 3 in response to chronic metabolic and respiratory acidosis.

The Na(+/)H(+) exchanger isoform 3 (NHE3) is essential for HCO(3)(-) reabsorption in renal proximal tubules. The expression and function of NHE3 must adapt to acid-base conditions. The goal of this study was to elucidate the mechanisms responsible for higher proton secretion in proximal tubules during acidosis and to evaluate whether there are differences between metabolic and respiratory acidosis with regard to NHE3 modulation and, if so, to identify the relevant parameters that may trigger these distinct adaptive responses. We achieved metabolic acidosis by lowering HCO(3)(-) concentration in the cell culture medium and respiratory acidosis by increasing CO(2) tension in the incubator chamber. We found that cell-surface NHE3 expression was increased in response to both forms of acidosis. Mild (pH 7.21 ± 0.02) and severe (6.95 ± 0.07) metabolic acidosis increased mRNA levels, at least in part due to up-regulation of transcription, whilst mild (7.11 ± 0.03) and severe (6.86 ± 0.01) respiratory acidosis did not up-regulate NHE3 expression. Analyses of the Nhe3 promoter region suggested that the regulatory elements sensitive to metabolic acidosis are located between -466 and -153 bp, where two consensus binding sites for SP1, a transcription factor up-regulated in metabolic acidosis, were localised. We conclude that metabolic acidosis induces Nhe3 promoter activation, which results in higher mRNA and total protein level. At the plasma membrane surface, NHE3 expression was increased in metabolic and respiratory acidosis alike, suggesting that low pH is responsible for NHE3 displacement to the cell surface.

Toxicity of carbon dioxide: a review.

The toxicity of carbon dioxide has been established for close to a century. A number of animal experiments have explored both acute and long-term toxicity with respect to the lungs, the cardiovascular system, and the bladder, showing inflammatory and possible carcinogenic effects. Carbon dioxide also induces multiple fetal malformations and probably reduces fertility in animals. The aim of the review is to recapitulate the physiological and metabolic mechanisms resulting from CO(2) inhalation. As smokers are exposed to a high level of carbon dioxide (13%) that is about 350 times the level in normal air, we propose the hypothesis that carbon dioxide plays a major role in the long term toxicity of tobacco smoke.

Complete intracellular pH protection during extracellular pH depression is associated with hypercarbia tolerance in white sturgeon, Acipenser transmontanus.

Sturgeons are among the most CO2 tolerant of fishes investigated to date. However, the basis of this exceptional CO2 tolerance is unknown. Here, white sturgeon, Acipenser transmontanus, were exposed to elevated CO2 to investigate the mechanisms associated with short-term hypercarbia tolerance. During exposure to 1.5 kPa Pco2, transient blood pH [extracellular pH (pHe)] depression was compensated within 24 h and associated with net plasma HCO3- accumulation and equimolar Cl- loss, and changes in gill morphology, such as a decrease in apical surface area of mitochondrial-rich cells. These findings indicate that pHe recovery at this level of hypercarbia is accomplished in a manner similar to most freshwater teleost species studied to date, although branchial mechanisms involved may differ. White sturgeon exposed to more severe hypercarbia (3 and 6 kPa Pco2) for 48 h exhibited incomplete pH compensation in blood and red blood cells. Despite pHe depression, intracellular pH (pHi) of white muscle, heart, brain, and liver did not decrease during a transient (6 h of 1.5 kPa Pco2) or prolonged (48 h at 3 and 6 kPa Pco2 blood acidosis. This pHi protection was not due to high intrinsic buffering in tissues. Such tight active cellular regulation of pHi in the absence of pHe compensation represents a unique pattern for non-air-breathing fishes, and we hypothesize that it is the basis for the exceptional CO2 tolerance of white sturgeon and, likely, other CO2 tolerant fishes. Further research to elucidate the specific mechanisms responsible for this tremendous pH regulatory capacity in tissues of white sturgeon is warranted.

Mechanisms of tissue hypercarbia in sepsis.

Intramucosal acidosis, that it is to say, an increased intramucosal-arterial PCO2 difference, is a common finding in clinical and experimental sepsis. Nevertheless, the physiologic significance of increases in tissue PCO2 is controversial, since CO2 can be generated by both aerobic and anaerobic biochemical processes. PCO2 can rise after buffering of protons produced in the hydrolysis of high-energy phosphate compounds by bicarbonate, or after the anaerobic production of acids, like lactate. In this case, it could represent tissue dysoxia. Alternatively, an increase in tissue PCO2 could denote hypoperfusion and diminished removal of the CO2 produced during the oxidation of pyruvate. In this last situation, aerobic metabolism might be preserved. In the present review, we discuss the physiologic mechanisms that determine tissue and venous hypercarbia during the three classic forms of hypoxia: stagnant, hypoxic and anemic hypoxia. The results of experimental studies suggest that tissue minus arterial and venoarterial PCO2 gradients primarily reflect alterations in tissue perfusion. These conclusions are further confirmed by a mathematical model of CO2 transport. In sepsis, however, tissue hypercarbia might develop despite normal or high cardiac output. This phenomenon has been initially interpreted as secondary to alterations in energetic metabolism, the so-called cytopathic hypoxia. Yet, new evidences show that the underlying mechanism to tissue hypercarbia in sepsis might be due to severe microcirculatory derangements. In summary, experimental results support the hypothesis that increases in tissue and venous CO2 are insensitive markers of tissue dysoxia, and merely reflect vascular hypoperfusion.

Hormonal regulation of alveolarization: structure-function correlation.

BACKGROUND: Dexamethasone (Dex) limits and all-trans-retinoic acid (RA) promotes alveolarization. While structural changes resulting from such hormonal exposures are known, their functional consequences are unclear. METHODS: Neonatal rats were treated with Dex and/or RA during the first two weeks of life or were given RA after previous exposure to Dex. Morphology was assessed by light microscopy and radial alveolar counts. Function was evaluated by plethysmography at d13, pressure volume curves at d30, and exercise swim testing and arterial blood gases at both d15 and d30. RESULTS: Dex-treated animals had simplified lung architecture without secondary septation. Animals given RA alone had smaller, more numerous alveoli. Concomitant treatment with Dex + RA prevented the Dex-induced changes in septation. While the results of exposure to Dex + RA were sustained, the effects of RA alone were reversed two weeks after treatment was stopped. At d13, Dex-treated animals had increased lung volume, respiratory rate, tidal volume, and minute ventilation. On d15, both RA- and Dex-treated animals had hypercarbia and low arterial pH. By d30, the RA-treated animals resolved this respiratory acidosis, but Dex-treated animals continued to demonstrate blood gas and lung volume abnormalities. Concomitant RA treatment improved respiratory acidosis, but failed to normalize Dex-induced changes in pulmonary function and lung volumes. No differences in exercise tolerance were noted at either d15 or d30. RA treatment after the period of alveolarization also corrected the effects of earlier Dex exposure, but the structural changes due to RA alone were again lost two weeks after treatment. CONCLUSION: We conclude that both RA- and corticosteroid-treatments are associated with respiratory acidosis at d15. While RA alone-induced changes in structure andrespiratory function are reversed, Dex-treated animals continue to demonstrate increased respiratory rate, minute ventilation, tidal and total lung volumes at d30. Concomitant treatment with Dex + RA prevents decreased septation induced by Dex alone and results in correction of hypercarbia. However, these animals continue to have abnormal pulmonary function and lung volumes. Increased septation as a result of RA treatment alone is reversed upon discontinuation of treatment. These data suggest that Dex + RA treatment results in improved gas exchange likely secondary to normalized septation.

Hypercapnic acidosis impairs plasma membrane wound resealing in ventilator-injured lungs.

The objective of this study was to assess the effects of hypercapnic acidosis on lung cell injury and repair by confocal microscopy in a model of ventilator-induced lung injury. Three groups of normocapnic, hypocapnic, and hypercapnic rat lungs were perfused ex vivo, either during or after injurious ventilation, with a solution containing the membrane-impermeant label propidium iodide. In lungs labeled during injurious ventilation, propidium iodide fluorescence identifies all cells with plasma membrane wounds, both permanent and transient, whereas in lungs labeled after injurious ventilation propidium iodide fluorescence identifies only cells with permanent plasma membrane wounds. Hypercapnia minimized the adverse effects of high-volume ventilation on vascular barrier function, whereas hypocapnia had the opposite effect. Despite CO2-dependent differences in lung mechanics and edema the number of injured subpleural cells per alveolus was similar in the three groups (0.48 +/- 0.34 versus 0.51 +/- 0.19 versus 0.43 +/- 0.20 for hypocapnia, normocapnia, and hypercapnia, respectively). However, compared with normocapnia the probability of wound repair was significantly reduced in hypercapnic lungs (63 versus 38%; p < 0.02). This finding was subsequently confirmed in alveolar epithelial cell scratch models. The potential relevance of these observations for lung inflammation and remodeling after mechanical injury is discussed.

Hypercapnic acidosis is protective in an in vivo model of ventilator-induced lung injury.

To investigate whether hypercapnic acidosis protects against ventilator-induced lung injury (VILI) in vivo, we subjected 12 anesthetized, paralyzed rabbits to high tidal volume ventilation (25 cc/kg) at 32 breaths per minute and zero positive end-expiratory pressure for 4 hours. Each rabbit was randomized to receive either an FI(CO(2)) to achieve eucapnia (Pa(CO(2)) approximately 40 mm Hg; n = 6) or hypercapnic acidosis (Pa(CO(2)) 80-100 mm Hg; n = 6). Injury was assessed by measuring differences between the two groups' respiratory mechanics, gas exchange, wet:dry weight, bronchoalveolar lavage fluid protein concentration and cell count, and injury score. The eucapnic group showed significantly higher plateau pressures (27.0 +/- 2.5 versus 20.9 +/- 3.0; p = 0.016), change in Pa(O(2)) (165.2 +/- 19.4 versus 77.3 +/- 87.9 mm Hg; p = 0.02), wet:dry weight (9.7 +/- 2.3 versus 6.6 +/- 1.8; p = 0.04), bronchoalveolar lavage protein concentration (1,350 +/- 228 versus 656 +/- 511 micro g/ml; p = 0.03), cell count (6.86 x 10(5) +/- 0.18 x 10(5) versus 2.84 x 10(5) +/- 0.28 x 10(5) nucleated cells/ml; p = 0.021), and injury score (7.0 +/- 3.3 versus 0.7 +/- 0.9; p < 0.0001). We conclude that hypercapnic acidosis is protective against VILI in this model.

Osteoblastic intracellular pH and calcium in metabolic and respiratory acidosis.

In vitro metabolic acidosis (Met) induces greater bone mineral resorption than respiratory acidosis (Resp). Met, but not Resp, inhibits osteoblasts which control many aspects of osteoclastic function. To determine whether at a similar decrement in extracellular pH, Met and Resp would induce different changes in intracellular pH (pHi) and/or intracellular calcium concentration ([Ca2+]i) of osteoblasts, we measured pHi and [Ca2+]i in an osteoblast-like rat osteosarcoma cell line (UMR-106). Cells were grown to confluence on glass slides and loaded with either 1.5 microM BCECF, for pHi, or 1.5 microM Fura-2, for [Ca2+]i, in control (Ctl; pH approximately 7.40, PCo2 approximately 40, [HCO3-] approximately 24) medium. The fluorescence ratio at excitation wavelengths of 502 and 440 nm was measured for pHi and at 340 and 380 nm for [Ca2+]i. Following a baseline scan in Ctl medium, cells were transferred to either Met (pH approximately 7.10, PCo2 approximately 40, [HCO3-] approximately 12), Resp (pH approximately 7.10, PCo2 approximately 80, [HCO3-] approximately 24) or Ctl conditions. Medium pH, PCo2 and [HCO3-] were held constant over the course of the experiment. Compared to Ctl, pHi was lower in Met (P < 0.001) and even lower in Resp (P < 0.001 vs. Met and vs. Ctl). These changes were maintained over the period of observation. Compared to Ctl, [Ca2+]i was higher in Met (P < 0.001) and even higher in Resp (P < 0.001 vs. Met and vs. Ctl) within 20 to 100 seconds. However, after 100 seconds [Ca2+]i was not different in the three groups.(ABSTRACT TRUNCATED AT 250 WORDS)

Stimulated osteoclastic and suppressed osteoblastic activity in metabolic but not respiratory acidosis.

When bone is cultured in acidic medium produced by a reduced bicarbonate concentration ([HCO(3-)]), a model of metabolic acidosis, there is greater net calcium efflux than when the same decrement in pH is produced by an increased partial pressure of carbon dioxide (PCO2), a model of respiratory acidosis. To determine the effects of metabolic and respiratory acidosis on bone cell function we cultured neonatal mouse calvariae for 48 h under control conditions (pH approximately 7.40, PCO2 approximately 41 mmHg, [HCO(3-)] approximately 25 meq/l) or under isohydric acidic conditions simulating metabolic (pH approximately 7.09, [HCO(3-)] approximately 12) or respiratory (pH approximately 7.10, PCO2 approximately 86) acidosis and measured osteoblastic collagen synthesis and alkaline phosphatase activity and osteoclastic beta-glucuronidase activity. Collagen synthesis was inhibited by metabolic (23.2 +/- 1.3 vs. 30.3 +/- 1.0% in control) but was not altered by respiratory (32.3 +/- 0.6) acidosis. Alkaline phosphatase activity was inhibited by metabolic (402 +/- 16 vs. 471 +/- 15 nmol P.min-1.mg protein-1 in control) but not altered by respiratory (437 +/- 25) acidosis. beta-Glucuronidase activity was stimulated by metabolic (1.02 +/- 0.06 vs. 0.78 +/- 0.05 micrograms phenolphthalein released.bone-1.h-1 in control) but not altered by respiratory (0.73 +/- 0.06) acidosis. Net calcium efflux in control was increased by metabolic (783 +/- 57 vs. 20 +/- 57 nmol.bone-1.48 h-1 in control) and by respiratory (213 +/- 45) acidosis; however, calcium efflux with metabolic was greater than with respiratory acidosis.(ABSTRACT TRUNCATED AT 250 WORDS)

Influence of hypoxic and hypercapnic acidosis on brain water content after forebrain ischemia in the rat.

OBJECTIVE: To determine whether hypoxia or hypercapnia superimposed on ischemia affects brain water content after ischemia. DESIGN: Prospective, randomized, controlled trial. SUBJECTS: Thirty-one male Wistar rats. INTERVENTIONS: The rats were assigned randomly into one of six groups: a) control; b) ischemia; c) ischemia combined with hypoxia; d) ischemia combined with hypercapnia; e) hypoxia; f) hypercapnia. Forebrain ischemia was induced for 5 mins by clamping both carotid arteries and inducing exsanguination. Either hypoxia or hypercapnia was induced until the arterial pH decreased to 7.0. The rats were decapitated after the protocol. MEASUREMENTS AND MAIN RESULTS: After the decapitation, the specific gravities of the neocortex, caudatoputamen, hippocampus, cerebellum, and midbrain were measured using a variable-density bromobenzene-kerosene column technique as an index of brain swelling. The specific gravities of the hippocampus and the neocortex were significantly lower in the ischemic group than in the control group. Specific gravities of the caudatoputamen and neocortex in the ischemia plus hypercapnia group, and specific gravities of the caudatoputamen, neocortex, and hippocampus in the ischemia plus hypercapnia group, were significantly lower than in the ischemia group. CONCLUSIONS: Cerebral water content increases more when ischemia is accompanied by hypoxia or hypercapnia than after ischemia alone. Hypoxic and/or hypercapnic acidosis during the periresuscitation period may be one of the causes of brain swelling after the resuscitation of patients after an anoxic-ischemic insult.

Adulto jovem esplenectomizado, com varicela e síndrome de angústia respiratória aguda fulminante / A splenectomized going patient, with chicken pox and fulminant acute repiratory distress syndrome

É apresentado o caso de uma paciente jovem esplenectomizada na infância após trauma abdominal e que aos 18 anos foi acometida por varicela, tendo apresentado síndrome de angústia respiratória aguda fulminante. Outras complicaçöes da varicela säo revistas, assim como alguns aspctos da síndrome de angústia respiratória aguda

Critical role of bicarbonate in calcium release from bone.

Calcium release from cultured bone is pH dependent; net calcium flux (JCa) from bone increases with decreasing pH. At a similar decrement in pH there is greater JCa when acidosis is produced by a low medium bicarbonate concentration ([HCO3-]), a model of metabolic acidosis (Met), compared with an increased medium PCO2, a model of respiratory acidosis (Resp). To separate the role of [HCO3-] from that of pH in inducing JCa we cultured calvariae for 3 h under three different neutral (pH approximately 7.4) isohydric environments [control (Ctl), fully compensated Met (C-Met), or fully compensated Resp (C-Resp)] and two different acid (pH approximately 7.1) isohydric environments (Met or Resp). During neutral pH (Ctl, C-Met, and C-Resp) there was JCa from bone during C-Met (decreased [HCO3-]), no net flux during Ctl (normal [HCO3-]), and JCa into bone during C-Resp (increased [HCO3-]); and JCa was correlated inversely with [HCO3-] (r = -0.824, n = 36, P less than 0.001). During acid pH there was greater JCa from bone during Met (decreased [HCO3-]) than during Resp (normal [HCO3-]); and JCa was again correlated inversely with [HCO3-] (r = -0.848, n = 22, P less than 0.001). JCa from bone during Met and Resp was greater than C-Met and C-Resp, respectively. The addition of the osteoclastic inhibitor salmon calcitonin did not alter the relative JCa results. Thus at a constant pH the magnitude of JCa from cultured neonatal mouse calvariae appears dependent on the [HCO3-]; the lower the [HCO3-], the greater the calcium efflux.(ABSTRACT TRUNCATED AT 250 WORDS)

Adaptation to respiratory acidosis in the turtle bladder.

The effect of in vivo respiratory acidosis for 4 and 48 hr was examined in the turtle bladder by placing turtles in hypercapnic chambers. Blood pH was significantly lowered and pCO2 was significantly elevated over control values both 4 and 48 hr, while blood bicarbonate was only increased after 48 hr. In vitro rates for H+ secretion determined by the reverse short-circuit current were significantly greater in bladders from 48 hr of respiratory acidosis than those of controls (27.3 +/- 2.7 vs 20.6 +/- 1.7 microA, P less than 0.05). In vitro rates for HCO3- secretion determined by pH stat were not altered. Fluorescence microscopy was used to study cell morphology. The number of carbonic anhydrase cells (corrected for the total number of cells) as determined by four different fluorescence stains (6-carboxyfluorescein, rhodamine 123, acridine orange, and 3,3'-diethyloxacarbocyaninine iodide) was increased both after 4 and 48 hr of respiratory acidosis. However, the number of HCO3(-)-secreting (beta subtype) carbonic anhydrase cells, determined by a probe for the anion exchanger, NBD-taurine, was not increased. In vitro 1% CO2 for 4 hr also resulted in an increase in H+ secretion and in the number of 6-carboxyfluorescein-positive cells, both of which could be blocked with SITS pretreatment. We conclude that CO2 changes the mucosal cells more toward the carbonic anhydrase phenotype, and that if NBD-taurine accurately identifies the beta cells, that the adaptation produces or recruits more alpha-carbonic anhydrase cells.

Effect of acute respiratory acidosis on two populations of intercalated cells in rat cortical collecting duct.

Recent studies suggest the presence of two populations of intercalated cells in the rabbit cortical collecting duct (CCD), one involved with hydrogen ion secretion and another that may play a role in bicarbonate secretion. The purpose of this study was to determine whether two populations of intercalated cells are present in the rat CCD and to establish their response to acute respiratory acidosis. Rats were studied during normal acid-base conditions and after 4-5 h of respiratory acidosis. In all animals light microscopy and transmission and scanning electron microscopy revealed two configurations of intercalated cells, type A with an extensive apical tubulovesicular membrane compartment and prominent surface microprojections and type B with a well-developed vesicular compartment and short sparse surface microprojections. By transmission electron microscopy, studs were present on the cytoplasmic face of the apical plasmalemma and tubulovesicular profiles of A cells. In respiratory acidosis there was a striking increase in apical microprojections and in the surface density of the apical membrane of type A cells similar to the response observed previously in intercalated cells in the outer medullary collecting duct (OMCD) studied under the same physiological conditions. No changes were observed in type B cells. Scanning electron microscopy revealed no change in the relative number of type A and type B cells in respiratory acidosis. We conclude that two distinct populations of intercalated cells exist in the rat CCD: type A, which resembles the intercalated cells in the OMCD, and type B. The response of type A cells to acute respiratory acidosis and the similarity between these cells and intercalated cells in the OMCD, which are believed to secrete hydrogen ion, suggest that the type A cells are involved in hydrogen ion secretion in the CCD.

Benzyl alcohol toxicity in a neonatal intensive care unit. Incidence, symptomatology, and mortality.

We unknowingly "screened" all NICU infants for elevated levels of serum benzyl alcohol (Bz-OH) over a three-month period. The fortuitous "screening" procedure resulted from the interference by Bz-OH with a routine blood CO2 assay used for all infants; validity was proved by (1) replication of the interference pattern with Bz-OH or benzoic acid; (2) confirmation of elevated benzoic acid levels in serum in two of the four screening positive infants tested but not in control infants, and (3) disappearance of the interference patterns when Bz-OH solutions were discontinued in affected infants. Screened Bz-OH-positive infants were compared to screened negative control infants, matched for weight (less than 1000 g), severity of RDS (on respirators), exposure to Bz-OH, and survival for longer than 48 hours. Intraventricular hemorrhage (IVH), metabolic acidosis appearing prior to IVH, and mortality were increased in Bz-OH-positive infants (P less than .05 in each case); hyperbilirubinemia and thrombocytopenia were not. Gasping respirations were not a major symptom. It is concluded that Bz-OH poisoning was a major cause of morbidity and mortality in NICU infants weighing less than 1000 g at birth during the three-month screening period. A retrospective review of patient records covering a 16-month period showed significant improvement in the survival rate of infants weighing less than 1000 g following the discontinuation of Bz-OH solutions.